You are on page 1of 226

0

GUIDANCE A N D NAVl GAT1O N

These p a p e r s a r e t o b e p r e s e n t e d by
t h e a u t h o r s a t W i e s b a d e n , Germany a n d
B r u s s e l s , Belgium i n June of 1965 under
t h e s p o n s o r s h i p o f NATO's A d v i s o r y Group
f o r Aerospace Research and Development.

R-500
SPACE NAVIGATION
GUIDANCE AND CONTROL

Volume 1 of 2
J U N E 1965

CAMBRIDGE 39, MASSACHUSETTS

by D r . C . S. D r a p e r
D r . W . Wrigley
D. G. Hoag
D r . R. H. Battin
J . E. Miller
D . A . Koso
D r . A . L. Hopkins
D r . W . E . Vander Velde

ACKNOWLEDGMENT

This r e p o r t w a s p r e p a r e d under DSR P r o j e c t 55- 238, sponsored by the


Manned Spacecraft Center of the National Aeronautics and Space Administration
through Contract NAS 9- 4065.
The publication of this r e p o r t does not constitute approval by the National
Aeronautics and Space Administration of the findings or the conclusions contained
therein.

It i s published only f o r the exchange and stimulation of ideas.

OCopyrighted by the Massachusetts I n s t i t u t e of Technology.


Published by the Instrumentation Laboratory of the
Massachusetts I n s t i t u t e of Technology.
Printed i n Cambridge, Massachusetts, U.S.A., 1965

ii

TABLE O F CONTENTS
Volume I of I1
Volume I
Part I

GUIDANCE - BASIC PRINCIPLES


Dr. C. Stark Draper
INTRODUCTION

Page
I- 3

I- 1

PROBLEMS O F GUIDANCE

I- 5

I- 2
I- 3

GEOMETRICAL ASPECTS O F GUIDANCE AND CONTROL


FUNCTIONAL REQUIREMENTS O F SYSTEMS AND THEIR
COMPONENTS FOR CONTROL AND GUIDANCE

1-15
I- 22

I- 4

STATEOFTECHNOLOGYOFCOMPONENTSFORCONTROL,
NAVIGATION AND GUIDANCE SYSTEMS

I- 37

I- 5

GYROSCOPE UNITS FOR REALIZATION O F GUIDANCE


SYSTEM R.EFERENCE COORDINATES

I- 42

I- 6

BASIC PRINCIPLES O F GYRO UNIT APPLICATIONS

I- 60

I- 7

SPECIFIC FORCE RECEIVERS

1-65

I- 8

INERTIAL SYSTEMS
BIBLIOGRAPHY

I- 70
1- 72

Chapter

Part I1
THE NAVIGATION, GUIDANCE, AND CONTROL OF A
MANNED LUNAR LANDING
David G. Hoag
11- 1

THE BACKGROUND AND THE PROBLEM O F SPACECRAFT


GUIDANCE, NAVIGATION, AND CONTROL

IT- 3

11- 2

GUIDANCE,NAVIGATION, AND CONTROL TASKS IN THE


APOLLO MISSION

11-21

11- 3

GUIDANCE,NAVIGATION, AND CONTROL INSTRUMENTATION


IN APOLLO
OPERATION MODES O F GUIDANCE, NAVIGATION, AND
CONTROL
APOLLO COMMAND MODULE BLOCK1
SPACECRAFT SAFETY CONSIDERATIONS OF GUIDANCE,
NAVIGATION, AND CONTROL

11- 45

11-4
11- 5

11- 67

11- 8 3

TABLE O F CONTENTS (Continued)

Part I11
EXPLICIT AND UNIFIED METHODS O F
SPACECRAFT GUIDANCE
Dr. Richard H. Battin
Chapter

Page
INTRODUCTION

111- 3
111- 5

111-2

ACCELERATED FLIGHT NAVIGATION


COASTING FLIGHT NAVIGATION

I11- 3

POWERED-FLIGHT GUIDANCE

111-2 7

MID- COURSE GUIDANCE

111- 39

BIBLIOGRAPHY

111- 51

111- 1

111- 4

Volume I1

111- 11

(Separately bound document)

Part IV
INERTIAL MEASUREMENT UNITS AND PULSE TORQUING
John F. Miller
IV- 1

THE APOLLO INERTIAL MEASUREMENT UNIT

IV- 2

THE PULSED INTEGRATING PENDULOUS ACCEZERONIETER


( PIPA)

IV- 3

THE COUPLING DATA UMT (CDU)


BIBLIOGRAPHY

IV- 3
IV-27
IV- 45
IV- 57

Part V
OPTICAL MEASUREMENTS AND NAVIGATIONPHENOMENA
D. Alexander Koso

v- 3
v- 5

INTRODUCTION

v- 1
v- 2
v- 3

NAVIGATION IN ORBIT
MID- COURSE NAVIGATION

V-25

THE APOLLO OPTICAL UNIT

V- 41

BIBLIOGRAPHY

v- 45

iv

TABLE O F CONTENTS (Continued)

Part VI
GUIDANCE COMPUTER DESIGN
Dr. Albert L , Hopkins, Jr.

INTRODUCTION

Page
VI- 3

VI- 1

CHARACTERISTICS O F GUIDANCE COMPUTERS

VI- 5

VI- 2
VI- 3

CHARACTERISTICS O F THE APOLLO GUIDANCE COMPUTER


MECHANIZED AIDS T O DESIGN AND PRODUCTION

VI- 1 3

VI- 4
VI- 5

GROUND SUPPORT EQUIPMENT


CONCLUSION

Chapter

BIBLIOGRAPHY

VI- 69
VI- 79
VI- 81
VI- 8 3

Part VI1
SPACE VEHICLE FLIGHT CONTROL
Dr. Wallace E. Vander Velde
INTRODUCTION
VII- 1
VI1- 2

POWERED FLIGHT CONTROL


COASTING FLIGHT CONTROL

VII- 3

ATMOSPHERIC FLIGHT CONTROL

VII- 3
VII- 5
VII- 40
VII- 51

PREFACE

The m a t e r i a l in t h i s book was assembled to support a s e r i e s of l e c t u r e s to be


given by the authors in Europe in June 1965, under the sponsorship of the Advisory
Group f o r Aerospace Research and Development, an agency of NATO.
The general subject of Space Vehicle Control Systems is the subject of discussion with particular application t o the present Manned Lunar Landing P r o g r a m .

The

man-machine interaction along with requirements of the mission a r e first described.


These mission requirements in t e r m s of specific hardware along with the performance
requirements and underlying r e a s o n s f o r choice a r e next explained. Lastly, the
theoretical background, the s y s t e m analysis and the derivation of the control functions
to integrate the hardware into a precision guidance, navigation and control system a r e
discussed.
lectures.

The book is organized into seven sections following the pattern of the

Section I provides historical background to the fundamental problems of guidance and navigation.

The basic physical phenomenon and associated instrument

techniques are discussed.


Section I1 continues w i t h background information going m o r e specifically into
the problems and approach of the guidance, navigation and control of the Apollo manned
lunar landing mission. This section i l l u s t r a t e s some of the basic philosophy and
approaches to the Apollo tasks, such a s the s u c c e s s enhancing decision t o provide
equipment that w i l l p e r f o r m all necessaryoperationson-boardand using a l l ground based
help when available.
Section I11 concerns in detail the analytic foundation f o r performing on-board
calculations f o r navigation and guidance, The achievement of a unified and universal

set of equations provides an economy in on-board computer p r o g r a m to p e r f o r m a l l the


various mission tasks.
Section IV c o v e r s in detail the mechanization of the inertial s e n s o r equipment
of the Apollo guidance and control system.

Section V provides the s a m e visability into the optical navigation s e n s o r s and


measurement techniques.

Section VI provides background and specific techniques in the mechanization of


on-board digital computers.

Application to the Apollo mission i l l u s t r a t e s s e v e r a l p r o -

b l e m s of i n t e r e s t such a s the method f o r providing reasonable and straightforward


astronaut data input and readout.
Section VI1 concerns the specific p r o b l e m s and solutions of vehicle attitude cont r o l under conditions both of rocket powered flight and the free- fall coast conditions.
The Apollo mission provides a diversity of examples of t h i s a r e a of technology in the
control schemes of the command and s e r v i c e module, the lunar landing vehicle, and
the e a r t h entry r e t u r n configuration.
The general p r o b l e m s of Space Navigation, Guidance, and Control r e q u i r e s a
g r e a t variety of discipline f r o m the engineering and scientific fields. The successful
completion of any one space mission o r phase of a space mission r e q u i r e s a t e a m effort
with a unified approach. Of equal importance a r e the software d e l i v e r i e s and p e r f o r m ance with the hardware.

This lecture series is an 3tke-t

t w i n t e g r a t e many of the

disciplines involved in creating successful and a c c u r a t e space vehicle control s y s t e m s ,


T h e s e l e c t u r e s r e p r e s e n t , on e v e r y o n e s part, a n interplay between equipment
and theory.

While in each c a s e emphasis may be o n one o r the other, in the whole

equal emphasis is applied.


A l l sections may be t r e a t e d a s s e p a r a t e entities however in the case of Section 111
through VI1 it is helpful to have the background of Section 11. T h e r e is c r o s s reference
between sections to avoid unnecessary duplication.

It is observed that the a u t h o r s have emphasized the Apollo mission and hardware
a s examples in t h e i r t r e a t m e n t of the subjects. This is partially because of t h e i r intimate familiarity with Apollo in the development work at the Instrumentatioll Laboratory of
MIT and partially because Apollo provides, in a n existing program, a n excellent example
in its multiple requirements and diversity of problems. Because Apollo is currently
under development, no p a r t i c u l a r attempt h a s been made to make reference only to the
latest configuration details, Indeed the a u t h o r s have utilized various s t a g e s of the
Apollo development cycle without specific identification in every c a s e a s they provide
the guidance, navigation, o r control technique example desired.
The a u t h o r s wish to e x p r e s s t h e i r appreciation t o NASA f o r the opportunity to
participate in the lecture series and f o r p e r m i s s i o n t o u s e m a t e r i a l from the r e s e a r c h
and development contracts NAS 9-153 and NAS 9-4065. They also recognize that t h i s
does not constitute approval by NASA of t h i s material. In addition, they wish to thank
the many m e m b e r s of M. I. T. ' s Instrumentation Laboratory; who a r e working on the
Apollo system, f o r t h e i r inspiration and generation of material.

PART I
GUIDANCE

- BASIC PRINCIPLES

Dr. C. S t a r k D r a p e r
Dr. Walter Wrigley

DR. CHARLES STARK DRAPER

Director, instrumentation Laboratory


Massachusetts Institute of Technology
Charles Stark Draper is Professor and Head of the Department ofAeronautics and
Astronautics and Director of the Instrumentation Laboratory. Dr. Draper is responsible for an extended curriculum of courses in the fields of instrument engineering and
fire control. These courses include regular instruction by the Instrument Section of
the Department of Aeronautics and also work leading to degrees for Navy and A i r Force
officers in armament and fire control. In addition, D r . Draper has written extensively
in the fields of instrumentation and control and has served a s consulting engineer to
many aeronautical companies and instrument manufacturers. Re holds a number of
patents for measurement and control equipment.
Dr. Draper was born in Windsor, Missouri on October 2,1901.

In 1922 he r e -

ceived a Bachelor of Arts degree in psychology from Stanford University. He then entered M.I.T. where he earned a B. s. in electrochemical engineering in 1926; an S.M.
without specification of department and an Sc. D in physics in 1938.
Dr. Draper is President of the International Academy of Astronautics and his
many memberships include the American Academy of Arts and Sciences, the National
Academy of Sciences, and the American Institute of Aeronautics and Astronautics. In
1965 President Johnson personally awarded the National Medal of Science to Dr. Draper
f o r his many engineering achievements f o r the defense of the country!

DR. WALTER WRIGLEY


Professor of Instrumentation and Astronautics
Massachusetts Institute of Technology
Walter Wrigley is Professor of Instrumentation and Astronautics, Educational
Director o f the Instrumentation Laboratory, and Acting Director of the Experimental
Astronomy Laboratory at the Massachusetts Institute of Technology. He has been active in the development of basic concepts f o r fire control and navigation systems and
has written extensively in these and related fields.
D r . Wrigley w a s born in Brockton,Mass., on March 26, 1913. He received his

Bachelor of Science degree in 1934 and an Sc. D. in Physics in 1941, both from M. I. T.
P r i o r to joining the Instrumentation Laboratory in 1946, D r . Wrigley served as
project engineer with the Sperry Gyroscope Company. He has served on numerous
military and civilian advisory boards. Dr. Wrigley is a registered Professional Engineer in Massachusetts, a member of the Scientific Advisory Board f o r the Chief of
Staff of the A i r Force, an Associate Fellow of the American Institute of Aeronautics
and Astronautics, and a member of several professional and honorary societies.
I- 2

Part I
GUIDANCE

BASIC PRINCIPLES

INTRODUCTION

Guidance is the p r o c e s s of collecting and applying information for the purpose


of generating maneuver commands to control vehicle movements.

In effect, this

p r o c e s s r e p r e s e n t s closure of the e s s e n t i a l control feedback branch that has to be


associated with s t r u c t u r e and propulsion in o r d e r f o r any vehicle system to operate
successfully.

Strong a i r f r a m e s and powerful engines can provide the capability for

flight, but without control to give stability, guidance t o determine proper paths and
to generate maneuver commands for realizing these paths, the m o s t sophisticated
and expensive craft a r e of no practical value.

The finest airplane will not begin to

s e r v e as a means of transportation until the pilot takes his seat and a s s u m e s the
functions of control and guidance. It is t r u e that in r e c e n t t i m e s the demands of
these functions have often gone beyond the abilities of human s e n s e s , human muscles
and the speed of human thought p r o c e s s e s .

Man m u s t now retain his hold on the

command of many situations through his invention, production and application of


inanimate devices that aid o r replace his own limited powers f o r direct action. The
collective brains of mankind a r e again demonstrating the ages-old truth that thought

is v e r y powerful among the f a c t o r s that determine p r o g r e s s .

Advances in the

technology of control and guidance have been substantial during the past two decades,
but these advances have generated much misunderstanding, controversy and often
strong opposition as the demands f o r attention and funding support have increased.
However, the spectacular r e s u l t s that have been achieved, particularly in the fields

of ballistic m i s s i l e s , submarines, satellites and space vehicles, have tended to


reduce this r e s i s t a n c e and to encourage the development of a technology essential t o
the advancement and perhaps the survival of our country.
A wide s p e c t r u m of possibilities f o r the future have already been revealed
by r e s u l t s now in the r e c o r d s , but essential decisions associated with a continuation
of work toward pioneering improvements in performance remain to be made in the
n e a r future.

Matters of national policy, strategy, tactics, economics, politics,

company profits and human emotions a r e so inter-mixed with basic physical laws

and technological developments that any significant clarifications of basic problems


associated with guidance and control a r e certainly helpful in forming plans for
constructive action. The authors of this paper hope t o provide some a s s i s t a n c e by
a discussion of basic principles, requirements, mechanization f e a t u r e s and natural
performance limitations of components and of s y s t e m s to meet the needs of m i l i t a r y
operations.

Representative numerical values f o r typical c a s e s a r e cited, but

specific results from particular equipments a r e not presented.

I- 4

CHAPTER I- 1

PROBLEMS O F GUIDANCE

The traditional method of directing the motion of a vehicle f r o m a port of


departure to a port of destination is based on the position and direction information
generated by navigation,

This situation is suggested by Fig, 1-1.

Because the

t e r m i n a l phases of many missions are made to depend upon direct contacts with
facilities at the destination, the a c c u r a c y required of navigation is in general not
v e r y great,

If navigation can bring a vehicle into an a r e a extending a few m i l e s

around the destination, its function h a s been accomplished.

F o r flights covering

not m o r e than a few hours it follows that performance inaccuracies not g r e a t e r than
one to t h r e e m i l e s for each hour of flight a r e often considered satisfactory f o r
navigation s y s t e m s .
Attacks on area targets with weapons able to cause destruction over a r e a s
.
purposes of such attacks can
s e v e r a l m i l e s in radius is suggested in Fig. 1 - 2 ~ ~ The
be s e r v e d by control and guidance s y s t e m s giving C E P ' s C i r c u l a r E r r o r Probability

(the radius in which half of a significant number of flights would t e r m i n a t e ) with the
o r d e r of one nautical mile,

This inaccuracy should be substantially independent of

range and t i m e of flight.


Any s c h e m e of navigation o r guidance that does not use direct contacts based

on either natural o r artificial electromagnetic (or acoustic) radiation, must depend


upon the identification of points on the e a r t h ' s s u r f a c e in t e r m s of m e a s u r e d distances
f r o m established bench m a r k s o r by means of angles between local gravity directions
with r e s p e c t to coordinate a x e s fixed in the earth.

The association of these directions

with points on a theoretical mapping s u r f a c e makes it possible to identify well-surveyed


positions on the e a r t h with inaccuracies somewhat l e s s than one-tenth of a nautical

mile, This m e a n s that a CEP of one-tenth nautical mile as the performance goal f o r
the instrumentation of navigation and guidance s y s t e m s is consistent with the m a p g r i d s
now available,

Effective attacks on many m i l i t a r y targets, such as bridges and

hardened m i s s i l e sites, which can be given m a p locations within one-tenth nautical


m i l e r e q u i r e inaccuracies around the a i m point of approximately the s a m e one-tenth
mile o r d e r of magnitude. Figure I-2b suggests the situation associated with a
ballistic attack on a hardened m i s s i l e site.

I- 5

Control, navigation and guidance always involve a reference coordinate


s y s t e m with axes having known working relationships with directions in the space
used for defining the essential path of motion.

This definition involves directions

with r e s p e c t to the r e f e r e n c e coordinates, and also distances and motion between


the guided entity and the destination o r target. In practice, the r e f e r e n c e coordinates
may be established in s e v e r a l ways and the essential distances m a y also be indicated
by various methods.
Figure 1-3 illustrates one of the simplest situations for guidance with one
airplane making an attack on another craft with guns.

The problem is f o r the

a t t a c k e r t o fly a path which causes projectiles f r o m his armament to s t r i k e the target.


His problem c e n t e r s around the line of sight to the t a r g e t with r e f e r e n c e coordinates
f o r maneuvers fixed in the attacking plane.

Usually roll, pitch and yaw axes fixed to

the a i r c r a f t would be instinctively chosen for judging direction and magnitude of


maneuvers. Success is achieved when the attacker flies SO that his gunfire destroys
the target.
Navigation and guidance present situations that a r e m o r e complex than the
circumstances of an a i r - t o - a i r duel because direct visual line-of-sight contact with
the destination is not generally possible, so that a reference space outside the moving
vehicle is n e c e s s a r y in o r d e r to describe positions and motions, When flight paths
a r e between points associated with the e a r t h ' s s u r f a c e it is n a t u r a l to use e a r t h ' s
coordinates established by north and the vertical f o r reference purposes.

It is also

reasonable to u s e radiation such as radio and r a d a r for determining distance and


direction.
Situations of many kinds appear, depending upon circumstances, but it is
possible to illustrate the principles involved in t e r m s of the diagrams of Figs. 1-4,

I- 5 and I- 6.
Figure 1-4 suggests the situation in which ground-based equipment having a
known orientation with r e s p e c t to e a r t h coordinates has artificial radiation contacts
with the moving vehicle. When these contacts a r e by pulse tracking, r a d a r distance
m e a s u r e m e n t s a r e direct and give position when combined with indications of
direction f r o m tracking signals,
Figure 1-5 illustrates the relationship of distances between points on the e a r t h ' s
s u r f a c e to angles between gravitational vectors a t the points in question, Measurements
of directions of gravity a r e generally associated with positions by the means of
celestial navigation procedures, In these methods the gravity vector angles f r o m lines
of sight to selected stars a r e corrected f o r e a r t h ' s rotation and then related to map
information,

It is noted on the figure that an inaccuracy of 60 seconds of a r c (one

minute of a r c ) gives a position inaccuracy of 6000 feet (one nautical mile), while
inaccuracies of 6 seconds of a r c and 1 second of a r c correspond to 600 feet and 100
feet respectively.
Figure 1 - 6 illustrates the indication of distance moved by a vehicle over the
s u r f a c e of the e a r t h by integration of signals from an a c c e l e r o m e t e r with i t s input
axis stabilized along the direction of vehicle motion,

One integration of these signals

A second integration gives


Assuming perfect orientation of the input axis during a one-hour

f r o m a given initial instant gives changes in velocity.


changes in position.

time of flight, an average a c c e l e r o m e t e r inaccuracy of 3 0 ~ 1 0 -e a~r t h gravity leads to


approximately 6000 feet inaccuracy in the indication of distance traveled, Under
s i m i l a r circumstances a c c e l e r o m e t e r inaccuracies of 3x10 - 6 and 0. 5x1 0 - 6 e a r t h
gravity produce approximately 600 feet and 100 feet respectively.
The u s e of a vehicle-borne a c c e l e r o m e t e r implies that the means to stabilize
the m e m b e r on which it is mounted be aligned with the direction of the e a r t h gravity
vectors which identify positions on the e a r t h ' s surface.

On the b a s i s of numbers given

in Fig. 1-6, initial alignment inaccuracies of 6 a r c seconds, 0. 6 a r c seconds and 0. 1


a r c second will mean position inaccuracies of 6000 feet, 600 feet and 100 feet
respectively,

Additional inaccuracies of s i m i l a r magnitude will accumulate if the

orientational r e f e r e n c e keeping the a c c e l e r o m e t e r input axis at right angles to the


local gravity directions drifts at average r a t e s that accumulate the given angular
inaccuracies,
It is convenient to e x p r e s s t h e s e drift r a t e s in t e r m s of e a r t h ' s r a t e f o r the
purposes of describing s y s t e m performance. E a r t h ' s r a t e called an "eru" unit is 15
degrees p e r hour o r 900 minutes p e r hour.

One-thousandth of e a r t h ' s r a t e (called

one milli- earth- rate- unit, one m e r u ) is thus one minute p e r hour (0. 015 degree p e r
hour) which corresponds to a position inaccuracy of about one nautical mile p e r hour,
This means that one-tenth nautical m i l e (600 feet) corresponds to one-tenth m e r u
(0.0015 degree p e r hour, o r 6 seconds p e r hour) while one-hundred feet corresponds
to 0.0167 m e r u (0.00025 degree p e r hour, 1 arc- second p e r hour).
It is to be noted that the numbers mentioned in the l a s t paragraph a r e only
rough approximations, Any mechanization would r e q u i r e higher performance from

its individual components in o r d e r to account f o r the interactions that inevitably


exist in complete s y s t e m s .
In s u m m a r y , it appears that radiation link inaccuracy is directly that of the
instrumentation used,

When gravitational directions a r e used to indicate positions,

a 60 arc- second e r r o r gives one nautical mile e r r o r , with the corresponding e r r o r


f o r 100 feet being one arc-second.

If one-hour flight t i m e is allowed to accumulate

these e r r o r s the stabilization drift e r r o r must be l e s s than 1 m e r u ( 0 . 015 degree


p e r hour) f o r one mile e r r o r in an hour while a drift of 0.0167 m e r u ( 0 . 00025 degree
per hour, 1 arc- second p e r hour) is required if not m o r e than 100 feet position
error is to be developed in one hour.
When signals f r o m an a c c e l e r o m e t e r with its input along the vehicle flight
path a r e used to generate position change data, one m i l e e r r o r in one hour needs an
-6
a c c e l e r o m e t e r inaccuracy of about 30x10
e a r t h gravity, while 100 feet e r r o r in
one hour needs a c c e l e r o m e t e r performance in the range of 0. 5 ~ 1 0 e- a~r t h gravity.

I- 8

-7

Lu

\
d

2
z

i;l

/'

//i

C E P O F ABOUT 1 MILE

AIM
POINT

\
\

"-\1-2"

/ APPROXIMATELY O N E ORDER

.".

OF M A G N I T U D E -LESS THAN
THE TOLERABLE ERRORS FOR
NAVIGATION

a. AREA TARGET GUIDANCE


H

..

c
0

\
C E P O F ABOUT 0.1 MILE (600 FT)

FOR EFFECTIVE RESULTS


FROM AVAILABLE WARHEADS

b. MILITARY TARGET GUIDANCE

Fig. 1-2

Guidance

"

+"

"

"
-.-

"

"
*
"
"
"

TARGET

/'

"" .

"_

.No-

LINE OF SIGHT FROM ATTACKER


TO TARGET

0
'

0
'

INITIAL VELOCITY OF PROJECTILE


DETERMINED RELATIVE TO
REFERENCE FRAME

-\//=
INITIAL PROJECTILE
VELOCITY

REFERENCE FRAME
ORIGIN AT CENTER-OF- MASS OF ATTACKER

PR I MARY Dl RECT ION

YAW AXIS

F i g . 1-3

LINE- OF-SIGHT FROM CENTER-OF-MASS


O F ATTACKER TO CENTER -OF - MASS OF
TARGET

Line of Sight Guidance

- Reference F r a m e

in Vehicle

DISTANCE TRAVELED IN PERIOD BETWEEN

DISTANCE FROM RADAR


DATA AT TIME t,

\DISTANCE
FROM RADAR
\
DATA AT TIME t2

RADAR PROVIDES DIRECT MEASUREMENT


- .
OF DISTANCE
~

DOPPLER RADAR PROVIDES RELATIVE


DATA (DISTANCE BY INTEGRATION]

ERROR IN DISTANCE EQUALS ERROR


IN MEASUREMENT

\
Fig. 1-4

AIRPORT

Position Change Information from Radiation Contact Measurements

ACCELEGOMETER OUTPUT GIVES


DIRECTLY - RESULTANT OF
ACCELERATION
AND GRAVITY
(GRAVITY ZERO IN
THIS SITUATION)
INTEGRATED ONCE- VELOCITY

STABILIZED MEMBER HOLDING


ACCELEROMETER WITH ITS
INPUT AXES PARALLEL TO
THE FLIGHT PATH

INTEGRATED TWICE- POSITION

,./L,cJl.
jJ

Fig. 1-6

APPROXIMATE TOLERABLE AVERAGE


ACCELEROMETER INACCURACIES FOR
VARIOUS DISTANCE INACCURACIES
.
AFTER 1 HOUR FLIGHT
ACCELEROMETER INACCURACY
DISTANCE INACCURACY
1 N. MILE
6000 FEET
3 0 ~ 1 0 -EARTH
~
GRAVITY
0.1 N . MILE
600 FEET
3x
EARTH GRAVITY
0.5 x
EARTH GRAVITY
0.0166 N.MlLE 100 FEET

Position and Velocity Changes from Integration of Accelerometer


Measurements Taken along Flight Paths

CHAPTER I- 2
GEOMETRICAL ASPECTS O F GUIDANCE AND CONTROL

F r o m the standpoint of basic geometry the problem of navigation and


guidance is that of commanding vehicles to move so that they r e a c h the vicinity of
destinations o r effectively hit pre- selected targets. In o r d e r for this p r o c e s s to be
at all possible a r e f e r e n c e space in which knowledge of relative location and motion
between the guided vehicle and i t s goal m u s t be available. With this knowledge in
hand it may be processed and compared with d e s i r e d location and motion to determine
indicated deviations f r o m which correction maneuver commands can be generated.
Geometrical reference space for navigation and guidance is not unique, but m a y be
chosen in many ways to be convenient for the problem under consideration.
Figure 1-7 illustrates a simple situation in which a guided vehicle and its
target are linked by a direct line of sight. This line is the geometrical entity that
determines the maneuvers c a r r i e d out by the pilot as he flies to the vicinity of his
target.

His own airplane acts f o r him a s the reference space f o r these maneuvers,

no outside body is involved. Here his reference space is naturally provided by the
e a r t h in good weather with the horizon and landmarks to give him the vertical and
north as a setting f o r the location of his destination.

In effect, the e a r t h supplies

t h r e e coordinate directions for judging angles and a ground-fixed point at the


destination f o r estimating distance and velocity along these axes,
Figure 1-8 shows how the situation changes when clear visual contacts with
the ground a r e lost because of night, weather o r t e r r a i n .

Under these circumstances

it is universal practice to u s e gyroscopic instruments responsive to gravity for


vertical indications and controlled by north- seeking devices for azimuth to establish
a s e t of coordinates f o r directional reference purposes.

It is significant to note that

these instruments by-pass the s t r u c t u r e of the vehicle by which they are c a r r i e d and
act as self- contained equipment in providing orientational reference coordinates.
With visual contacts eliminated, artificial radiation links at radio and r a d a r
frequencies a r e used to establish distances and directions f r o m known ground-based
stations to the vehicle.

Thus, directional information in e a r t h coordinates is

provided by both on-board and remote equipment while indications of distance come

Figure 1-9 illustrates the situation that exists when both visual contacts f o r
judging orientation and artificial radiation links f o r indications of position a r e not
available,

The devices providing orientational r e f e r e n c e information must be

improved to have s e v e r a l o r d e r s of magnitude s m a l l e r e r r o r r a t e s than the


orientational reference needed f o r the situation of Fig. 1-8.

By giving the geometrical

reference m e m b e r an initial alignment accurately related to e a r t h coordinates at a


known point, changes in location m a y be indicated by effectively carrying out the
double integration of a c c e l e r o m e t e r output, o r by following changes in direction of
the e a r t h ' s gravity vector. This means, in effect, that e a r t h coordinates for a
selected time and place must have been t r a n s f e r r e d to a self-contained system aboard
the guided vehicle. With the good equipment performance n e c e s s a r y to provide
information continuously and accurately on e a r t h space directions and properly mounted
a c c e l e r o m e t e r s with output signals processed by computers, a self-contained s y s t e m
will indicate location and velocity with r e s p e c t to its point of departure.

A system
which operates in this way is called an INERTIAL GUIDANCE SYSTEM, Systems of
this kind a r e universally used in ballistic m i s s i l e s and in submarines when guidance
without outside contacts is important. Inertial guidance s y s t e m s for s e r v i c e over
the e a r t h a r e implemented in many ways, but the necessity r e m a i n s for an accurate,
continuously available, self- contained geometrical reference related in a known way
to the external space in which guidance is to be performed. A number of typical
mechanizations a r e described briefly in a l a t e r section of this report.
When guidance is considered for vehicles to operate not in the n e a r vicinity
of the e a r t h but in regions f o r which the earth, moon, planets and s t a r s effectively
approach m a s s points, coordinates aligned in e a r t h space lose their usefulness.
Rather, it is n e c e s s a r y to employ geometrical r e f e r e n c e coordinates associated with
s t a r s and planets,

For example, celestial s p h e r e coordinates, or some other

directions such as a line directed toward the sun, m a y be s e t into an inertially


stabilized r e f e r e n c e m e m b e r , Figure I- 10 shows the essential f e a t u r e s of the Apollo
Guidance System which is to be used on manned flights to the moon and return. This
system must deal reliably and accurately with some fifteen o r twenty different
problems of guidance ranging f r o m e a r t h launching through e a r t h orbit, mid- course
to the moon, moon orbit, moon landing and r e t u r n to .orbit, and finally through space
t o landing at a pre- selected point on the earth.
An inertial m e m b e r with provisions f o r either manual o r automatic alignment
with e a r t h coordinates o r celestial coordinates is used as the geometrical r e f e r e n c e
during acceleration phases of the trip.

Visually o r automatically established lines of

sight to known s t a r s and to landmarks on the e a r t h and the moon a r e used as the b a s i s
f o r determining location and velocity during mid- course flight. Data from these

observations a r e processed by a digital computer which supplies both navigational


information and guidance commands f o r s y s t e m operation, which m a y be completely
automatic as self- contained equipment, completely manual o r partially automatic
with monitoring by on-board human pilots, o r with remote monitoring by ground-based
s u p e r v i s o r s , through radio and r a d a r links. It is probable that the Apollo Guidance
Systems which have already been conceived, designed, built, tested and delivered
will be useful models f o r Space Guidance Systems of the future,

I- 17

rI

I- 19

c,

.d

c,

cd

a
I
H

1- 20

Ea,

c,

(I]

h
rn

0
4

1-2 1

CHAPTER I- 3
FUNCTIONAL REQUIREMENTS O F SYSTEMS AND THEIR
COMPONENTS FOR CONTROL AND GUIDANCE

Figure I- 11 r e p r e s e n t s the essential geometrical function of stabilization in


t e r m s of right-angled three- coordinate axis s y s t e m s associated with the s p a c e s
concerned, A s e t of stabilization reference coordinates is established by some
instrumental means o r by direct visual contact to a space either identical with o r
related to the space in which vehicle motion is to be guided. An e s s e n t i a l duty of
the control s y s t e m is to cause a x e s fixed to the vehicle s t r u c t u r e to r e m a i n
continually close to the stabilization r e f e r e n c e axes, and a l s o to keep the vehicle
velocity vector substantially identical with a stabilization r e f e r e n c e velocity vector
which has a direction and a magnitude established in s o m e way with r e s p e c t to the
r e f e r e n c e space in which the vehicle path is defined.
Figure 1- 12 r e p r e s e n t s maneuver, a second control system function in which
the stabilizafion reference coordinates and the stabilization r e f e r e n c e motion a r e
changed with r e s p e c t to the vehicle path r e f e r e n c e space in the ways n e c e s s a r y to
accomplish m i s s i o n s ,

The control system inputs that s e r v e this purpose a r e maneuver

commands generated f r o m plans, p r o g r a m s , feedback data, environmental data, and


other information by a guidance system. The nature and functions of this s y s t e m a r e
illustrated by the diagram of Fig. I- 1 3 .
F i g u r e s I- 14 and I- 15 suggest the control and guidance situation that existed
during the e a r l y days of manned flight. Without a pilot to complete the informationhandling feedback loop of control and guidance, an airplane was completely u s e l e s s .
Plans and p r o g r a m s were stored in the m a n ' s brain, stabilization r e f e r e n c e s and
airplane conditions were noted by human s e n s e s and processed in the pilot's mind to
,generate maneuver commands that were applied to the airplane control l e v e r s by his
hands and feet.
F i g u r e s I- 1 6 and I- 1 7 illustrate the control and guidance situation that
commonly exists today in jet aircraft. Human pilots continue to be used in the control
and guidance s y s t e m s but their s e n s e s a r e greatly extended by radio, r a d a r and many
instruments, t h e i r muscle f o r c e s a r e boosted by s e r v o power and their ability to
solve complex problems rapidly is extended by computers.

I- 22

A l l of t h e s e appurtenances

certainly irnprove the effectiveness of control and guidance, but the pilot's position
a s an "on-line" component in both the control and guidance loops means that his
limitations in ability to solve complex problems rapidly and p r o p e r l y handle
situations requiring too rapid responses, s e t boundaries to the possible performance
of the overall system.
F i g u r e s I- 18 and I- 1 9 suggest the circumstances that exist in rocket powered
vehicles that, because of limited payload capacities, one-way missions, hostile
environments, and s e v e r e p r o g r a m s must operate with self-contained automatic
control and guidance s y s t e m s ,

The absence of restrictions imposed by human

limitations m a k e s it possible to realize ballistic m i s s i l e s and other vehicles with


capabilities well beyond those in which men provide control and guidance functions

as "on- line" components.


Figures 1-20 and 1-21 illustrate the situation that exists in the control and
guidance s y s t e m of a manned vehicle to operate in the astronautical regions above
the e a r t h ' s atmosphere and beyond the e a r t h ' s gravitational field. The control and
guidance s y s t e m s a r e automatic with orientational and translational r e f e r e n c e s
provided by an inertially stabilized m e m b e r , and a s e t of t h r e e a c c e l e r o m e t e r s
rigidly mounted on this m e m b e r with input axes s e t in an orthogonal configuration.
A telescope and a space sextant with their line-of-slight directions adjustable and

t r a n s f e r a b l e through a computer to the inertial reference m e m b e r give information


f o r correcting reference member alignment. Observations by the human pilot o r by
automatic optional tracking also suppry data for a computing s y s t e m to calculate
positions in space and to generate correction maneuver commands.
Flight condition data displayed by the automatic control and guidance s y s t e m
to the human pilot provide the information f o r monitoring s y s t e m operation, A s e t
of controls forming the operation mode s e l e c t o r and optional command s y s t e m make
it possible f o r the pilot to determine the mode in which the overall s y s t e m works. He
may select any s o r t of configuration f r o m full automatic, in which he only o b s e r v e s
operation, to completely manual, in which he acts to close servo-loops by continued
on-line operation.
F i g u r e s 1- 22 and 1-23 suggest the complete configuration used by Apollo in
which a ground-monitoring s y s t e m connected by radio, r a d a r and possibly visual up
and down links to the flight vehicle.

The earth- based s y s t e m a c t s a s an information

collecting and monitoring branch in parallel with the on-board pilot monitor.

- whether

o r not they
a r e accepted in any given c a s e depends on operating doctrine and the circumstances
of particular cases.
Information and suggestions may be sent to the space vehicle

,-i
,-i

a
Lu
A

U
I

Lu

>

I- 2 4

DESIRED EQUILIBRIUM PATH I


/

DESIRED STABILIZATION,
VELOCITY VECTOR/

J
+

I
I

/\
EXISTING
EQUILIBRIUM PATH

REFERENCE
COORDINATES

VELOCITY VECTOR
IN VELOCITY VECTOR
-COMMANDED CHANGE IN
STABILIZATION REFERENCE

/ /

N
cn

/ /

REFERENCE COORDINATES

4
DESIRED
STAB1 L IZATION
REFERENCE PATH

__ ----

GE IN STABILIZATION
REFERENCE PATH
--
G T A B I L I Z A T I O N RE F ERE N C E PATH

MANEUVER IS THE PROCESS OF


CHANGING THE STABILIZATION REFERENCE
COORDINATES AND VECTOR VELOCITY SO THAT
VEHICLE ORIENTATION AND VELOCITY LEAD
TO MISSION S U C C E S S
Fig. I- 12

Maneuver

a,
V

c
cd

5
M
,-I

I- 26

1- 27

VEHICLE

ENVIRONMENTAL FORCES

CONTROL
ACTUATION
COMMANDS

VEHICLE ORIENTATION,
LOCATION, VELOCITIES
ACCELERATION, ETC.

F i g . I- 15

Guidance and Control System - Human Operator

.I

1- 29

I
LI

ENVIRONMENTAL FORCES

CONTROL
ACTUATION
COMMANDS

VEHICLE

GEOMETRICAL QUANTITIES
VEHICLE ORIENTATION,
LOCATION, VELOCITIES
ACCELERATION,ETC.

H
I

ENVl WNMENTAL QUANTITIES

MANEUVER
COMMANM

Fig. 1-17

GUIDANCE LOOP

Guidance and Control System - Human Operators Senses and Power Extended

Fig. I- 18

Automatically Guided Flight Vehicle

1-3 1

STABILIZATION LOOP
~

I COUPLING

GEOMETRICAL QUANTITIES
VEHICLE ORIENTATION,
LOCATION, VELOCITIES
ACCELERATION,ETC.

CONTROL INFORMATION SYSTEh

SYSTEM

CONTROL
COMMANDS

INFORMATION
PROCESS1NG
CONTROL
SYSTEM
INFORMATION
SIGNALS

H
I

W
N

ENVIRONMENTAL QUANTITIES

Fig. I- 19

Guidance and Control System - Automatic Operator

.I

i-

il

1- 33

Q
r

t
LA

1-34

1- 3 5

ENVIRONMENTAL

FORCES

VEHICLE MOTION

VEHICLE

ENVIRONMENTAL QUANTITIES

r-""

i I
I I

STABILIZATION LOOP

OPERATIONAL MODE SELECTION


OPTIONAL CONTROL INFORMATION COMMANDS

II

ENVIRONMENTAL QUANTITIES

_"

CONTRbL
ACTUATION
COMMANDS

'

CONTROL INFORMATION

-? t

-VEHICLE

FEEDBACK
QUANTITIES

SYSTEM

CONTROL INFORMATION DISPLAY SIGNALS

MANEUVER
COMM ANDS

GUIDANCE LOOP

/GEOMETRICAL
QUANTITIES
VEHICLE ORIENTATION,
LOCATION, VELOCITIES,
ACCELERATION. ETC.

OPERATIONAL MODE SELECTION.


OPTIONAL OPERATOR GUIDANCE COMMANDS

""_

GUIDANCE AND NAVIGATION

SYSTEM

cn

GROUND MONITOR SYSTEM

TRANSMITTER
DATA PROCESSING
AND COMPUTER SYSTEM

Fig. 1- 23

1 .

Control and Guidance System with Ground Based Monitor

CHAPTER I- 4
STATE O F TECHNOLOGY O F COMPONENTS FOR CONTROL,
NAVIGATION AND GUIDANCE SYSTEMS

1. Radiation links which a r e generally employed to implement friendly

environments have the function of establishing contacts f r o m t r a n s m i t t e r s to


r e c e i v e r s , transponders and reflecting bodies,

These links provide communications

by voice, by t e l e m e t r y and by signals of other kinds,

Pulsed and continuous wave

r a d a r indicate line-of-sight directions and distances. L a s e r s and ordinary s e a r c h lights also offer powerful radiation links, particularly for satellites and space
vehicles.
In the c u r r e n t state of radiation link technology which allows determination of
distances within a few f e e t , the links themselves do not impose limiting restrictions

on the performance of navigation and guidance s y s t e m s as far as distance


A,

m e a s u r e m e n t s a r e concerned.

The ability of radiation links to determine line-of-

sight directions with inaccuracies less than one milliradian is adequate for the needs
associated with navigation and guidance,
Carefully surveyed ground station s i t e s with commonly available indicators
of the vertical provide e a r t h reference coordinates of such high a c c u r a c y that
radiation beam orientations m a y be taken as substantially perfect. On the other hand,
radiation links established by airborne t r a n s m i t t e r s generally have adequate ability
to m e a s u r e distances, but have r e s t r i c t e d ranges due to limitations on size, weight

F o r these r e a s o n s
the accuracy of directional tracking is not so good a s that obtained f r o m ground
stations. A generally m o r e s e v e r e limitation of air- borne radiation link equipment
is introduced by inaccuracies of reference coordinate equipment which will always be
g r e a t e r than the corresponding inaccuracies of t r a n s m i t t e r s rigidly fixed to the earth.
This s o u r c e of reduced performance m a y be serious, and, in any case, m u s t be given
careful consideration in evaluating the errors of any particular system.
and power consumption of the geometrical stabilization m e m b e r ,

Computing Systems receive essential information and c a r r y out the


mathematical p r o c e s s e s n e c e s s a r y to generate required outputs, The problems
solved range f.rom trigonometric transformations to the determination of position
2.

and velocity f r o m a c c e l e r o m e t e r output signals.

In t e r m s of a rough analogy,

computers perform the s a m e functions that the brain of a pilot provides when a human
being a c t s as the on-line data processing component in navigation and guidance
equipment.
The c u r r e n t technology of computers, particularly those based on digital
operations, is so well developed that units of ample capacity, speed and reliability
with reasonable sizes and weights and power consumptions a r e available for u s e in
guidance systems.

Improvements in a l l essential computer f e a t u r e s including

r e s i s t a n c e to environmental interference e f f e c t s a r e now in p r o g r e s s ,

It is c e r t a i n

that mechanization of computing functions is not now and w i l l not in the future be a
limiting factor on navigation and guidance systems.
3.

Engineering problems associated with angle sensing servomechanisms,

data transmission, mechanical design, etc., have current solutions that a r e


generally satisfactory with advances c e r t a i n t o appear in the n e a r future.

Except for

c e r t a i n special situations these f a c t o r s do not limit the performance of equipment for


navigation and guidance.
4.

Guidance s y s t e m coordinates r e l a t e d in a known way to r e f e r e n c e directions

of the space in which the d e s i r e d path of the guided vehicle is defined a r e e a s i l y established when rigid or optical connections t o the ground a r e available.

When the guidance

s y s t e m is vehicle-borne i t s n e c e s s a r y r e f e r e n c e coordinates must be established with


a known relationship to external space and maintained with this relationship defined
during the p r o g r e s s of guided flight. This situation is suggested by the diagram of
Fig. 1- 24 with the guidance s y s t e m coordinates initially established before flight with
known geometrical relationships t o the flight path space r e f e r e n c e coordinates.
F o r the purposes of guidance, the system coordinates must continue t o

provide a geometrical r e f e r e n c e that accurately r e p r e s e n t s the flight path space as


the vehicle moves t o complete i t s mission. Because mechanical connections a r e
impossible and radiation links between the vehicle and the flight path r e f e r e n c e space
generally absent, the only possibility for realizing satisfactory guidance system
r e f e r e n c e coordinates l i e s in the use of inertial principles.

P r o p e r l y applied, these

principles make it possible to mechanize a member which either r e m a i n s nonrotating with r e s p e c t to inertial space, or moves in a quantitative way w i t h r e s p e c t
to this space.

Details of arrangements to accomplish such r e s u l t s a r e discussed in

the next section of this paper.


Current technology is easily able to provide guidance system r e f e r e n c e
coordinates representing flight space coordinates within one minute of a r c for each
hour of operation.

The a r c - second inaccuracy r e q u i r e d by military guidance

for hard t a r g e t s is m o r e difficult to achieve, but is feasible with proper attention t o

1-38

design and production of components,

The principles available, typical arrangements

and performance realized a r e discussed in the next section.


Specific f o r c e r e c e i v e r s , the devices commonly called a c c e l e r o m e t e r s , a r e
the only available means for on-board sensing translational vehicle motion when
5.

radiation links with the environment a r e not available,

A commonly used configuration

of specific force r e c e i v e r s is t o mount t h r e e units rigidly to the geometrical


r e f e r e n c e m e m b e r with their input axes aligned with the guidance r e f e r m c e axes,
This arrangement is suggested in the diagram of Fig. 1-25,
Signals f r o m the t h r e e specific force r e c e i v e r s r e p r e s e n t the resultant
components of gravity force and inertia reaction force along each of the t h r e e axes.
With the geometrical relationships of these axes to the flight space reference axes
known, calculations based on t h e s e signals make it possible for the computer to
generate output signals giving the changes in vehicle location and velocity
occurring after the s t a r t of system operation.
Assuming perfect alignment of the guidance system reference coordinates
with the flight space reference coordinates, and perfect computer operation, e r r o r s

in indicated location and motion a r e due to imperfections in specific force r e c e i v e r


performance.

Performance matching the requirements of navigation is e a s y to

realize, while m i l i t a r y guidance for hard point targets is within the capabilities of
today's advanced technology. The mechanizations that afford these r e s u l t s a r e
described in a l a t e r section.

1-39

GUIDANCE SYSTEM REFERENCE COORDINATES


INITIAL CONFIGURATION ESTABLISHED
WITH RESPECT TO FLIGHT PATH R E F ERENCE COORDINATES
I

,
1
f

n
I

GUIDANCE SYSTEM REFERENCE COORDINATES MAINTAINED


DURING FLIGHT WITH A KNOWN RELATIONSHIP TO FLIGHT
PATH SPACE REFERENCE COORDINATES

GEOMETRICAL
REFERENCE
MEMBER

I
I

'\
FLIGHT PATH SPACE REFERENCE
COORDINATES

k
Fig. 1-24

Relationship of Guidance System Reference Coordinates t o Flight Path


Reference Coordinates

SPECIFIC FORCE RECEIVERS


( A CCELEROMETE R S )

GEOMETRICAL REFERENCE
INPUT AXES HELD BY
GEOMETRICAL REFERENCE
IN K N O W N DIRECTIONAL
RELATIONSHIPS WITH
FLIGHT SPACE REFERENCE
COORDINATES

FLIGHT PATH SPACE REFERENCE


COORDINATES

Fig. 1-25

Specific F o r c e Receiver (Accelerometer) System

CHAPTER I- 5
GYROSCOPIC UNITS FOR REALIZATION O F GUIDANCE SYSTEM
REFERENCE COORDINATES

Reference coordinates f o r vehicle-borne guidance s y s t e m s with satisfactory


performance m u s t fulfill the general requirements summarized in Fig, 1-26. The
b a s i c functions a r e : 1) to provide continuously available mechanism reference
directions; 2) to provide accurate control of the r e f e r e n c e directions with r e s p e c t
to a selected external reference space in response to command signals; 3) to
provide angular output signals that accurately r e p r e s e n t deviations of c a s e fixed
reference directions f r o m the mechanism reference directions.

In addition to these

e s s e n t i a l performance characteristics, practical instruments m u s t be reliable, of


reasonable s i z e and weight, and be available at acceptable cost.
Gyroscopic principles m a y be applied to mechanize practical instruments f o r
providing guidance system reference directions,

The theory involved is associated

with applications of the Newtonian Laws of Mechanics to rapidly spinning s y m m e t r i c a l


rotors,

It is relatively simple to discuss this theory in t e r m s of vectors representing

rotational quantities in accordance with the commonly used "right hand'' conventions
that are summarized in Fig. 1-27,

The central idea is that a rotational quantity such

as angular velocity o r angular momentum may be described by a vector along the


axis of rotation with its length proportional to the magnitude of the quantity, and the
head of i t s a r r o w related to the direction of rotation by the "right-hand s c r e w rule,
Figure 1-28 suggests the basic operating principle of a gyroscopic element.
When the gyroscopic element definition condition of constant spin velocity exists,
Newton's Law of dynamics leads to the conclusion that a torque applied to the r o t o r
a t right angles to the spin axis causes the angular momentum vector to change i t s
orientation with r e s p e c t to inertial space with an angular velocity of precession
proportional to the magnitude of the torque and having a s e n s e that always t u r n s the
angular momentum vector toward the torque vector.
One way of using the gyroscopic element to realize mechanism reference
directions f o r guidance s y s t e m purposes is to s e t up an angular momentum vector,
and then carefully to reduce all torque components on the r o t o r to z e r o about any
axis at right angles to the spin axis,

In this torque- less condition the angular

I- 42

'I

momentum vector maintains an orientation with r e s p e c t t o inertial space that is completely determined by the direction about which the spinning torque originally built

up the angular momentum of the rotor.

Once the spin angular momentum vector direc-

tion is established it becomes useful for r e f e r e n c e purposes only through the medium
of signals that r e p r e s e n t angles between the spin axis and reference directions fixed
to the c a s e within which the r o t o r spins.

Figure 1-29 suggests the situation that exists

when a c a s e is rigidly fixed t o some b a s e that may have any a r b i t r a r y changes in


orientation with r e s p e c t t o inertial space.

The position of the c a s e with r e s p e c t to the

r e f e r e n c e direction can be indicated in t e r m s of the two angles between perpendicular


lines fixed t o the case, and the spin axis,

Signals representing these two angles may

be p r o c e s s e d by a computer t o give information about the position of the c a s e with


r e s p e c t to the spin axis in t e r m s suitable for any particular problem of control and
guidance.
The situation that is generally of p r a c t i c a l interest r e q u i r e s information on
the orienta.tion of the c a s e with r e s p e c t to some specified external reference space.
It is obvious that any single direction such as a spin axis can not specify angular
positions of the c a s e about t t e spin direction so that a second gyroscopic element with
i t s spin axis having some projection a t right angles to the f i r s t spin axis is required
for any complete indication of c a s e orientation in t h r e e dimensional space. When two
spin axis directions a r e available for mechanism r e f e r e n c e purposes, the common
position of two gyro unit c a s e s rigidly connected together gives all the needed
information.

It is t o be noted that the interpretation of this information depends on

the continuous operation of a computer t o c a r r y out complex trigonometrical


calculations.

The relationships of s p h e r i c a l trigonometry a r e such that computations

generally give r e s u l t s of varying a c c u r a c y as c a s e orientations change by l a r g e angles


with r e s p e c t t o spin axis directions. This fact, coupled with difficulties of achieving
satisfactory signals f r o m l a r g e components of case- to-gyro r o t o r angles in the
fractional a r c second region means that the "large deviation angle" configuration
illustrated in Fig. 1- 29 is not suitable for the mechanism r e f e r e n c e coordinate
indications of high performance guidance systems.
Problems of mechanism r e f e r e n c e coordinates associated w i t h sensing and
transformation of l a r g e angles a r e usually eliminated, so far as gyro units themselves
a r e concerned, by mounting the unit c a s e s on a member having t h r e e degrees of
angular freedom with r e s p e c t t o the base by which it is carried, and providing power
drives of some kind to overcome inaccuracy-producing torques due t o inertia and
friction. Figure I- 30 suggests the essential f e a t u r e s of such a n arrangement f o r
two-degrees-of-freedom with angular deviation signals between the e a s e and the spin
a x i s direction used to energize gimbal torquer drives through the operation of
electrical servo- s y s t e m s not show11 in the diagram.

I- 43

With proper s e r v o designs, the

case-to-spin axis angular deviations m a y be maintained s m a l l ; in practice to the


o r d e r of one arc- second by using good angle s e n s o r s and tight servo-loops.

only s m a l l angles a r e involved and accurately matched sensitivities a r e unimportant,


difficulties associated with accurate signals a r e greatly reduced from those involved
in collecting data on l a r g e angles.

Because
t'

Combinations of the signals from two perpendicular

c a s e directions in a plane a t right angles to the spin axis a r e required t o command


proper responses from the two torquers.

Relatively low performance r e s o l v e r s on

the gimbal axes a r e sufficient to s e r v e the needs of s e r v o loop control, It is


unnecessary to include a s e p a r a t e trigonometric computer in the system.
The diagram of Fig. 1-30 illustrates a single r o t o r arrangement providing
two-degree-of-freedom isolation of a controlled m e m b e r (the case) f r o m base motion.
Any complete s y s t e m r e q u i r e s three-degree-of- freedom angular isolation.

In practice

this situation is v e r y often m e t by an arrangement like that suggested in the diagram of


Fig. 1-31. In this figure, t h r e e gyro units, each with one degree-of-freedom, a r e
shown on the inertial reference package instead of the two that would be needed if gyro
elements like that of Fig. 1-30 were used.

F o r the situation illustrated, the nature of

the gyro units is i m m a t e r i a l s o long as they provide the function of sensing and
representing angular deviations in t e r m s of usable signals,
Gyro units of many types have been conceived and a few have been reduced to
successful practice.

Strong discussions of relative m e r i t f o r various mechanizations

continue and a r e not likely to be settled until working equipment is tested under
operational conditions. However, an understanding of the patterns in which fundamental
principles m a y be applied is s u r e l y helpful f o r effective evaluations of performance data
f r o m t e s t results. The discussion that follows is intended to help with this
understanding by describing the f e a t u r e s and problems associated with typical c l a s s e s

of gyro units.
~ l gyro
l
units a r e designed around a relatively s t r o n g component of angular

momentum with its direction rigidly fixed to some m e m b e r which has freedom to move
within the instrument case. For the instruments that have proved to be successful in
the present state of technology, this angular momentum is generated by a spinning
r o t o r of s o m e kind. Figure 1- 32 illustrates the simplest arrangement in which a
single moving part, a r o t o r having a generally s p h e r i c a l f o r m , is supported on f o r c e s
generated by electromagnetic or electrostatic fields S O configured that the resultant
damping f o r c e s acting on the r o t o r a r e verylow. This c h a r a c t e r i s t i c makes it possible

f o r a r o t o r forced into rotation by eddy-current motor acting to continue coasting with


a high velocity spin for considerable periods of time such as days, weeks o r months
after the driving torque has been removed.

When a s p h e r i c a l r o t o r runs within a

spherical c a s e in the arrangement suggested by Fig. 1-32, the frictionless support not

I- 44

only provides a spin bearing but also allows for complete angular freedom of the c a s e
with r e s p e c t to the spin axis.

Sensors for angles between r e f e r e n c e s fixed to the c a s e

and the spin axis supply signals that yield orientational reference information after
processing by a computer,
The inviscid field supported r o t o r gyro unit is attractive because of its
simplicity but s t a r t - u p of spin is an awkward p r o c e s s requiring considerable time
with the spin direction determined by the orientation of the c a s e during the t i m e the
spin torque is acting.

Nutation, i. e. , an oscillatory change in direction of the

angular momentum vector is generated during the starting operation and must be
damped out.

The inviscid field supported r o t o r arrangement does not lend itself to

the controlled application of torque components for adjusting the direction of the
angular momentum vector, a circumstance which makes it practically impossible to
align the spin axis direction directly and accurately with external r e f e r e n c e c a s e
coordinates.

This means that auxiliary means to provide special positioning of the

c a s e with r e s p e c t to external coordinates must be used, The required equipment


tends to be cumbersome and difficult to use. This circumstance makes it unlikely
that inviscid field supported s p h e r e gyro units will be as satisfactory for the purposes
of high performance guidance s y s t e m s a s other units that allow self-alignment of the
system in which they operate,
Gyro units with inviscid field supported spherical r o t o r s also present certain
other difficulties,

These problems s t e m f r o m the inaccessibility of spinning r o t o r s

for balancing and other adjustments while the complete gyro unit is in operation with
all the environmental conditions adjusted to those of operational use.

Another m a t t e r
of basic importance that r e m a i n s to be resolved is that of the effects of vibration and
acceleration on a gyroscopic system with substantially z e r o damping.

It is certainly

v e r y d e s i r a b l e to m e a s u r e environmental effects and to determine overall system


performance as soon a s possible.
Because gyro r o t o r behavior can not be refined by m a s s changes made
directly on spinning s p h e r e s , it is not possible to refine gyro performance by
adjustments. In practice the operation of each individual sphere m u s t be calibrated
in combination with a computing system to determine performance coefficients that
can be applied during operation to reduce imperfections in behavior by calculation
r a t h e r than through adjustment o r compensation of the mechanism.

P r a c t i c a l problems of design, engineering, production and operation for gyro


units may be simplified by separating the various functions that must be provided
within a gyro unit in ways that allow each aspect of performance to be given individual
adjustments and compensation with a minimum of coupling effects that lead to

I- 45

inaccuracies in operation. Figure 1-33 suggests the basic f e a t u r e s of the typical


two-degree-of-freedom gyro unit with:
a) m g u l a r momentum provided by the rotation of a wheel-like r o t o r spinning,
c a r r i e d by an inner gimbal through shaft and journal bearings that may be ball, roller
o r hydro-dynamic with gaseous o r fluid lubricant.
b) Spin angular velocity sustained (no coasting in operation) with constant
speed by a continuously acting motor.
c) Two-degrees of angular freedom with r e s p e c t to the case provided by
gimbals (fluid, ball o r r o l l e r supported.)
d) Generation of angular deviation signals, r e s t r i c t e d to s m a l l magnitudes,
by reception of spherical displacements of the case with respect to the spin axis.
e) Accurate changes in angular momentum orientation with r e s p e c t to
inertial space by direct response to command input for gimbal t o r q u e r s .
f ) Balance adjustments available during unit operation by means of threaded
nuts on the two gimbals, These adjustments make it possible to approach the ideal
condition of gyro unit insensitivity to gravity and acceleration by adjusting the center
of m a s s so that it approaches coincidence with the point of support provided by the
gimbals,
In s o m e designs this mechanical support m a y be supplemented by flotation
f o r c e s provided by liquid within the clearance between hermetically sealed thin shell
gimbals.
When its base is mounted on a s t r u c t u r e which rotates with r e s p e c t to inertial
space, the two-degree-of-freedom gyro unit gives output signals which r e p r e s e n t spin
axis angular deviations about axes perpendicular to the spin axis, from a reference
position of the case, This reference position is determined by the orientation of the
c a s e in which the angle output signal has its null level.

Command signals to the

t o r q u e r s make it possible to change the reference orientation as desired without any


need f o r taking base orientation into account.

This possibility of directly relating

gyro unit angular momentum to an external space reference direction gives the torqued
two-degree-of-freedom gimbal supported gyro unit a considerable advantage over
inviscid field supported s p h e r i c a l r o t o r units,
The gimbal supported two-degree-of-freedom gyro unit overcomes some of
the difficulties stemming from the multiple function c h a r a c t e r i s t i c s of inviscid field
s p h e r i c a l r o t o r supports. Adjustments can be made during operation by gimbal
balancing adjustments, indefinite operating periods a r e achieved, accurate control
of angular momentum directions is made possible and various other results of
practical importance a r e attained. However, the difficulties associated with
accurately maintaining coincidence between the point of support provided by two

I- 46

gimbals coincident and the center of m a s s of an articulated structure, limit the


quality of performance available from the two-degree-of-freedom gyro unit and
the difficulty of manufacture for units of even medium

considerably i n c r e a s e
performance levels.

The output of deviations in t e r m s of a conical angle subject

to the coupling effects that tend to accompany the precession and nutation of a twodegree-of-freedom gyro r o t o r is also troublesome when inaccuracies in the region
of fractional arc- seconds a r e desired.
Some difficulties of two-degree-of-freedom units a r e reduced when the
mechanical gimbal s y s t e m is replaced by a s p h e r i c a l gas bearing arrangement which
allows both r o t o r spin action and angular freedom between the spin reference
direction and the case.

Balancing problems still exist and the gas bearing is always

subject to sharply defined upper limits of r e s i s t a n c e to shock and vibration, but many
practical gyro units using this mechanization a r e in operational use.
All two-degree-of-freedom gyro units a r e essentially untorqued f o r the

purposes of sensing angular deviations.

This means that the obtainable resolution in

t e r m s of angular velocity components about axes of sensitivity for the unit depends on
the angle defined by the minimum usable output f r o m the signal generator. This
generally corresponds to angles so l a r g e that detection of s m a l l components of e a r t h ' s
r a t e (in the region of one a r c second p e r hour) is generally not practical,
Single-degree-of-freedom gyro units with the basic f e a t u r e s illustrated in
Figs. I- 34 and I- 35 make it possible to realize practical gyro units with the
c h a r a c t e r i s t i c s required of angular deviation sensing instruments f o r guidance
s y s t e m s able to reliabIy provide the low C E P range needed f o r h a r d point m i l i t a r y
targets,

The design philosophy involved is the direct antithesis of the single moving

p a r t philosophy of the inviscid field supported s p h e r e gyro unit in which refining


adjustments to the r o t o r a r e impossible during operation, and accurate direct
alignment of angular momentum axis to external reference space is not available. In
the floated integrating single-degree-of-freedom gyro unit each function is carefully
separated f r o m others, and with the exception of dynamic balancing for the rotor, m a y
be refined toward ultimate performance with the complete gyro unit in normal operation.
As suggested by Figs, 1-34 and 1-35:
a) Angular momentum is provided by a r o t o r with i t s spin sustained

indefinitely by a driving motor.

The spin axis bearings which support the r o t o r f r o m

the single gimbal m a y be either ball bearings o r hydrodynamic journal bearings


lubricated by air. With good design and manufacture both types have demonstrated
high performance and life t i m e s of many thousands of hours.

Gas bearings consume

somewhat m o r e power at starting and in operation than ball bearings.

I- 47

Gas bearings

a r e much m o r e liable to damage than balls when subjected to torques before their
full supporting power has been developed during start- up, and they a r e also m o r e
vulnerable to catastrophic failure under either steady o r vibratory high accelerations.
It a p p e a r s that gas bearings a r e suitable for environments of limited severity, while
ball bearings a r e adaptable to wider ranges of environmental conditions. Both types
a r e now in u s e and can be applied in single- degree- of-freedom gyro units at the
preference of the designer.
b) Single-degree-of-freedom motion is provided by a chamber enclosing the
gyro r o t o r which is l a r g e l y supported on the flotation p r e s s u r e gradients built up by
gravitational and inertia reaction f o r c e s in a dense, highly viscous fluid contained in
the clearance volume between the float and the hermetically sealed case, In operation,
the t e m p e r a t u r e s of the solid p a r t s and the fluid a r e closely controlled so that the
buoyancy support r e m a i n s substantially constant. The s m a l l remaining imperfection
in flotation is effectively reduced to z e r o by single axis magnetic support units at
either end of the float which is thus suspended within the c a s e without even the
slightest rubbing contact between solid parts.
c) Complementing the buoyancy and magnetic supporting f o r c e s a r e f o r c e s
generated by hydrodynamic f o r c e s and torques generated when the heavy fluid is forced
to flow between p a r t s of the clearance space.

The resultant support minimizes

distorting s t r e s s on the gimbal because of the distributed nature of the loads that a r e
acting, The overall result is a s y s t e m effectively immune t o the mechanical effects
of acceleration, vibration and shock within selected design ranges.
d) The use of high viscosity fluids for gimbal support means that high level
d r a g f o r c e s a r e developed by motion of the float within the case. The f o r c e s not only
provide support during dynamic conditions, but also develop a d r a g torque about the
gimbal freedom axis of the float.

This torque is proportional to the angular velocity

of the float with r e s p e c t to the case about this axis which is carefully made at right

angles to the spin axis of the gyro rotor.

Under gyroscopic principles a rotation of

the an'gular momentum vector about the input axis which is at right angles to the gimbal
axis, and also the spin axis causes the gyro r o t o r to e x e r t a torque on the gimbal

~axis. This torque


about its axis of freedom which, f o r this reason, is called the output
is absorbed by the accelerational inertia reaction of the float, and by viscous d r a g in

the fluid.

The inertia reaction torque causes the gyro unit to exhibit a t i m e constant

with the o r d e r of one thousandth second, while the viscous d r a g torque causes the float
angular velocity within the case to be proportional to the c a s e angular velocity with
r e s p e c t to inertial space about the input axis. The result is that the float output angle
with r e s p e c t to the case is proportional to the angle turned through by the case with
r e s p e c t to inertial space about the input axis. This action leads the name "integrating
gyro unit" for instruments with the f e a t u r e s of Fig. 1-34.

I- 48

Because of the absence of rubbing friction between solid p a r t s and the


utilization of viscous d r a g a s a p r i m a r y factor in operation, the proportionality
between input angle and float output angle is effectively perfect over the operating
range of a few minutes of a r c down to a lower limit that

js

s u r e l y less than one

thousandth of an a r c second.
e) In operation, the single-degree- of- freedom integrating gyro unit is
suitable for u s e only under circumstances in which gimbal deflection angles a r e
limited to a few seconds of a r c .

This condition is favorable f o r signal generator

designs which can be constructed to give v e r y low null signals and high sensitivity
outputs when connected to feasible electronic circuits. Signals defining less than 0. 01
a r c seconds a r e achievable with carefully designed g e n e r a t o r s and good electronics.
f ) AS shown in Fig. 1-34 balance adjustments for the float have the f o r m of
nuts on s c r e w s attached to the float,

By providirg means f o r turning these nuts f r o m

outside the c a s e with the unit in full operation, it becomes possible to place the center
of m a s s on the output axis so accurately that the total effects of unbalance torques
under one e a r t h gravity m a y be reduced to the level of one m e r u o r less.
g) Flexible leads carefully selected f o r low h y s t e r e s i s and substantially floated
in the suspension fluid a r e used to c a r r y power to the gyro r o t o r within the gimbal
float. By careful design and the u s e of a refining adjustment accessible f r o m outside
the case, the effects of power lead torque m a y be reduced to a level of one m e r u o r
less.
With single axis operation and accurately controlled centralization of the
moving element, torque g e n e r a t o r s a r e feasible in which the output is v e r y closely
proportional to the electrical input,

This c h a r a c t e r i s t i c makes the single- degree- of-

freedom integrating gyro unit a v e r y useful and flexible component f o r applications of


many kinds.
The f e a t u r e s described above in g e n e r a l t e r m s coupled with an absolutely
n e c e s s a r y c a r e f u l control of tempe.rature, r o t o r power, electrical excitation,
mechanical mounting, connectors, etc. , make it possible to reduce gyro performance
uncertainties to levels between 0 . 0 1 and 0 , 0 0 1 m e r u .

By p r o p e r compensation a n d / o r

correction of various b a s i c effects which a r e measurable by inspection techniques,


gyro performance substantially identical with the uncertainty levels m a y be achieved
in p r a c t i c a l operation.

I- 49

O U T P U T 7 7 A
SIGNAL

0
0

DEFINE A
MECHANISM
REFERENCE DIRECTION
(TO ORDER O F
ARC- SECOND
FRACTION)

GENERATE A SIGNAL REPRESENTING ANGULAR


DEVIATIONS O F A C A S E REFERENCE DIRECTION
FROM THE M E C H A N I S M REFERENCE DIRECTION
WITH RESOLUTION TO ARC- SECOND FRACTION
tie. 1/10 "* ~ / I O O O )

@ ROTATE WITH RESPECT TO A N EXTERNAL REFERENCE


w

SPACE AS QUANTITATIVE RESPONSE TO C O M M A N D


SIGNAL

0,

COMMAND SIGNAL

@ RELIABLE CONTINUOUS OPERATION FOR INDEFINITE PERIOD

Fig. 1-26

@
@

(SEVERAL YEARS AT LEAST)


REASONABLE POWER C O N S U M P T I O N
REASONABLE SIZE A N D W E I G H T

REASONABLE COST

General Requirements f o r Instrumental Components t o Mechanize Guidance


System Reference Directions

VECTOR POINTS I N OlRECTlON


OF RIGHT-HANO THUMB WHEN
FINGERS LINE UP AS INDICATED
WITH SENSE I N WHICH THE ROTATIONAL QUANTITY ACTS

l(ap)W(sp)

kg')

W(ap)

vector agular tomentum

of ntor(rpinning about
a axis of symmetly)
momat of inertia of
mtiting body about
spin axis
spin vdocity
(vector agular
velocity about
spin axis)

VECTOR WlNTS ALONG AXIS OF


SPIN IN ACCORDANCE WITH RIGHTHAND RULES
LENGTH OF VECTOR IS PROPORTIONAL TO MAGNITUDE OF ANGULAR
MOMENTUM

SPIN

AXIS

ROTATING BOY (mtor)


a ) Right-hmd rules for the relationship betwaen a rotatianal qrontity m d the vector
representing the qumtity

W - symbol for r a b l a r

w-

SENSE OF
ROTATION

velocity vector
symbol for speed of
mtation

REFERENCE LINE

ANnrt

7 ROTATED LINE
rL

PLANE CONTAINING RL
AN0 NORMAL TO PLANE
CONTAINING RL AN0 r L

~~~~~~~
BODY

b ) Vecior representutian of on angular velocity

6 - w e t o r reprewting perpen-

e ) Angle represented as a dihedral male betwen p l m e s

VECTOR POINTS ALONG TORQUE


AXIS ACCORDING TO RIGHT-HANO
RULES
LENGTH OF VECTOR IS PROPORTIONAL TO MAGNITUOE OF TORQUE

dicular arm Mwm brce

REFERENCE LINE

SENSE OF
TORQUE
TORQUE AXIS

For thla diagram, thetnm "torque" haa a mmninp Identical wlth the word "couple"
aa u u d In convntlonal mechanlca.
CJ

..-... ". -

PLANE
RI

OIHEORAL ANGLE
BETWEEN PLANES
OF ROTATION ACCORDING TO
RIGHT-HAND RULES
LENGTH OF VECTOR IS PROPORTIONAL TO SPEED OF ROTATION

AXIS OF ROTATION

d) Vector representation for the mgular momentum of a spinning rotor

Vector'repesentation of a torque

Fig. 1-27

f) Angle represented as a mtotional vector

Vector Conventions for Rotational Quantities

1-51

E!

0
e
(3

zZ
z-

a
v)

Q
I-

O
a
a

3
v)

I- 52

i"

a,

0
.ri

a
0

[I]

k
U

FIRST REFERElNCE DIRECTION


F A
FIXEL\ rI nU LH
SE

FOR INSTRUMENTS)
GYROSCOPIC SPHERE
SECOND REFERENCE DIRECTION
FIXED TO CASE (IN PRACTICE
REPRFSENTED BY SIGNAL
GENERATED BY MEANS
IVOT S H O W N )

FIRST CASE ORIENTATiON ANGLE


- CPlhl AYIC fFCTARI ICUFC A

I
-4

REFERENCE"EQUAT0R" O N ROTOR

"<"\

'-

SECOND CASE ORIENTATION ANGLE

INDICATIONS OF ORIENTATION OF THE CASE WITH RESPECT TO THE SPIN AXIS DIRECTION DEPEND ON
COMPUTER PROCESSING OF ANGLE SIGNALS BY RELATIONSHIPS O F SPHERICAL TRIGONOMETRY.

FOR COMPLETE DEFINITION O F INDICATED CASE ORIENTATION WITH RESPECT TO INERTIAL SPACE TWO
GYROSCOPIC ELEMENTS WITH DIFFERENT SPIN AXIS DIRECTIONS ARE REQUIRED.
ARBITRARY CASE ORIENTATION INDICATIONS WITH INACCURACIES WITH THE ORDER OF O N E ARC-SECOND
ARE DIFFICULT BECAUSE OF THE HIGH SIGNAL GENERATOR ACCURACY AND COMPUTER
PERFORMANCE REQUIRED.
Fig. I- 29 Illustrative Reference Direction Mechanization with Untorqued
Gyroscopic Elements and Computed Indications of Case Orientation
from Direct Rotor-Case Angle Signals

SECOND CASE REFERENCE

FIRST CASE REFERENCE

ANGULAR DEVI AT ION


ABOUT FIRST REFERENCE
DIRECTION AXIS

READOUT OF ANGLES
ANGULAR DEVIATIONS ABOUT THE
CASE REFERENCE DIRECTION AXES
(IN THE PLANE OF T H E ROTOR SPIN
EQUATOR WHEN THE ANGULAR
DEVIATIONS ARE ZERO) ARE R E CEIVED BY SIGNAL GENERATORS
GIVING OUTPUTS PROPORTIONAL TO
DEVIATIONS OF THE REFERENCE
DIRECTIONS FROM PERPENDICULARITY TO THE SPIN AXIS.
THESE SIGNALS APPLIED AS SERVO-DRIVE INPUTS
TO GIMBAL TORQUERS ACT TO KEEP THE CASE
Is

ANGULAR DEVIATION
ABOUT SECOND REFERENCE
DIRECTION AXIS

Fig. 1-30

AS A CONTROLLED MEMBER IN ALIGNMENT


WITH THE SPIN AXIS.
ANY SATISFACTORY SERVO-SYSTEM MAINTAINS
THE ANGULAR DEVIATIONS SMALL- ONE SEC.
OF ARC REPRESENTS REASONABLE PERFORMANCE.

Illustrative Two-Axis Mechanism Reference System Based on


Servo-Driven Gimbals and the Untorqued Gyroscopic Element

Fig. 1-3 1 Illustrative Arrangement of Three-Degrees-of-Freedom Servo-Driven


Gimbals t o Isolate Mechanism Reference Members f r o m B a s e Motion

0
I<

Lu

-I

(3

Z
0 c3

(3

Z
a:

a:
c

<

Lu

-<

Z
a
v)

-1

Lu

CL

Lu
A

a:
Lu

LL

z
2z

I-

CL

0
&

m
I

2
m

YZ

I-

32
nZ

<
e

ln

I- 56

6 0

SPIN MOTOR
INNER GIMBAL

T W O DEGREE OF ANGUL.AR
SENSOR

TORQUER

FEATURES

SPIN MOTOR SUSTAINS ROTOR


SPEED CONSTANT INDEFINITELY
SPIN BEARINGS LIMIT ROTOR TO
ROTATION ABOUT SPIN AXIS FIXED
TO INNER GIMBAL
TWO DEGREES OF ANGULAR FREEDOM FOR ANGULAR MOMENTUM
VECTOR PROVIDED BY GIMBALS

H
I

rn
-3

ANGLE DEVIATION SIGNALS REPRESENT COMPONENTS OF SINGLE ANGLE

NOTE: ROTOR A N D ITS BEARINGS MAY BE


ENCLOSED IN A HERMETICALLY SEALED
CAN A N D SUBSTANTIALLY SUPPORTED
BY FLOTATION IN A FLUID CONTAINED
WITHIN A N OUTER CASE
Fig. 1-33

SENSITIVE TO ANGULAR MOTION


ABOUT ALL AXES AT RIGHT ANGLES
TO SPIN AXIS

TORQUERS ON GIMBALS MAKE IT


POSSIBLE TO CHANGE ORIENTATION
OF SPIN AXIS AS A QUANTITATIVELY
ACCURATE RESPONSE TO COMMAND
SIGNALS

Basic F e a t u r e s of the Gimbal Supported Two-Degrees-of-Freedom Gyro Unit

COMPLETEGYRO
ON MOUNT

UNBALANCE
COMPENSATOR

POWER

UNBALANCE
COMPENSATOR
ADJUSTMENT

TORQUE GENERATOR
AND MAGNETIC
SUSPENSION
UNIT

H
I

ul

02

UNIT

Fig. 1-34

POWER
LEAD
BAFFLE
ASSEMBLY

Basic F e a t u r e s of Single-Degree-of-Freedom Floated Integrating Gyro Unit

GIMBAL
BEARING

SIGNAL
GENERATOR

(ab)

OUTPUT AXIS

GYROSCOPIC
ELEMENT
(e.)

\
GIMBAL
TORQUE
GENERATOR

SPIN REFERENCE AXIS

CASE
DAMPER

SPIN AXIS
(SA)

A[(ca)+im)l

FIXED T O

INPUT AXIS
SPINS
FIXED T O CASE
AT RIGHT ANGLES
TOTHEOUTPUT
AXIS AND T H E
SPIN R E F E R E N C E
AXIS

3. T H E SYMBOL A[(rer)4cnprd)] R E P R E S E N T S T H E
ANGLE A MEASUREDFROMTHEREFERENCE
DIRECTION ((ref) IN T H E SUBSCRIPT) T O T H E
COMPARED LlRECTlON (lcmpd) IN T H E SUBSCRIPTI.

NOTES: 1. POSITIVE SENSES S H O W BY T H E ARROWS ARE


CHOSEN SO THAT (IA), (SRA), AND (OA) FORM A
RIGHT-HANDED SYSTEM.
2.THEGYROUNITTEMPERATURECONTROLPOWER
I S S U P P L I E D T O A MOUNTING BLOCK A D A P T E D
T O R E C E I V E T H E GYRO UNIT CASE. T H E FLOW
OFPOWERISCONTROLLEDBYTHEDIU*IPER
T E M P E R A T U R E SETTING.

R E F E R E N C E DIRECTIONCOMPARED DIRECTION
ANGLE

CASE

- (sa) - (t9) -

DAMPER

- (dmp) -

T H E STRUCTURE T H A T GIVES S U P P O R T FOR


T H E INTERNAL WORKING PARTS O F T H E G Y R O
UNIT ENCLOSES T H E P A R T S AND CARRIES
PROI~ISIONSFOR EXTERNALCONNECTIONS OF
A L L KINDS.
COMPONENT FOR RECEIVING I N P U T SIGNALS
AND PRODUCING CORRESPONDING OUTPUT
T O R Q U E A P P L I E D T O T H E GIMBAL ABOUT T H E
OUTPUT AXIS.
SUB S Y STEM RECEIVING ANGULAR VELOCITY O F
T H E GIMBAL WITH R E S P E C T T O T H E CASE AS
ITS I N P U T AND PRODUCING AS O U T P U T A RETARDING T O R Q U E ACTING ON T H E GIMBAL
ABOUT T H E O U T P U T AXIS WITH A MAGNITUDE
PROPORTIONAL T O T H E MAGNITUDE O F T H E
ANGULAR VELOCITY O F T H E GIMBAL WITH RES P E C T T O T H E CASE.

GYRO UNIT
SIGNAL
GENERATOR
GIMBAL

Fig,

I- 59

- (gu) - RTHIS
ENTITY MADE U P O F T H E COMPONENTS
E P R E S E N T E D IN THIS DIAGRAM AND A L L T H E
- (so) -

- (pim) -

ADDITIONAL P A R T S NECESSARY F O R A SINGLE


PACKAGETOCARRYOUTTHEFUNCTIONSOF
A GYRO UNIT.
COMPONENTFORRECEIVINGTHEANGLEOF
T H E SPIN AXIS WITH R E S P E C T T O T H E CASE AS
INPUT AND PRODUCING A CISRRESP9)NDING SI@
N A L THAT SERVES AS T H E O U T P U T SIGNAL
FROM T H E G Y R O UNIT.
STRUCTURE CARRYING T H E BEARINGS FOR T H E
SPINNING ROTOR O F T H E GYROSCOPIC ELEMENT,
ROTORSFORTHETORQUEGENERATORANDSIG
NALGENERATOR P A R T O F T H E D A M P E R FLOAT
SEALS AND S T R U ~ T U R E , BALANCE ADJU~TMENTS,
S T O P S , PIVOTS, ETC.

CHAPTER I- 6
BASIC PRINCIPLES O F GYRO UNIT APPLICATIONS

Single-degree-of-freedom gyro unit s y s t e m s as components a r e combined with


many m o r e devices to produce a coordinated overall result in guidance systems. The
p a r t i c u l a r function of any single gyro unit is to translate the resultant of rotations
about i t s input axis with r e s p e c t to inertial space and command inputs to i t s torque
generator into a resultant signal that r e p r e s e n t s the angular deviation of the float
f r o m the position for which the output signal has i t s null level, In effect, the output
signal r e p r e s e n t s the angular deviation of the case about the input axis f r o m a
reference position established by the null level of the signal, Command inputs to the
torque generator have the effect of rotating the reference position with an angular
velocity proportional to the signal.
Figure 1-36 is an illustrative pictorial schematic diagram in t e r m s of a single
axis s y s t e m suggesting the basic f e a t u r e s and operating principles of a typical gyro
unit

- servodriven controlled m e m b e r combination,

F o r the arrangement of this

figure which provides functions s i m i l a r to those of the geometrical reference m e m b e r


of an inertial guidance s y s t e m , it is assumed that the command input is z e r o except,
perhaps, for s m a l l compensations f o r calibrated imperfections of the particular gyro
unit involved.
When, f o r any reason, the base of the servo- drive moves so that the gyro unit
is rotated away f r o m i t s reference orientation about the input axis, o r a torque is
imposed f r o m any other s o u r c e , a gyro output signal is generated.

Through slip rings

this signal is applied a s input to the s e r v o - d r i v e s y s t e m which applies torque to the


controlled m e m b e r to force it back to the orientation f o r which the gyro unit case h a s
i t s reference position, As a result of this continued action of the s e r v o in overcoming
disturbing torques, the c a s e effectively holds i t s r e f e r e n c e position about the input
axis no m a t t e r how the base m a y move.

Operation of this is typical of geometrical

stabilization.
The servo- drive- gyro unit combination of Fig. 1-36 provides s e v e r a l functions
for inertial guidance s y s t e m s . The broad n a t u r e s of these functions a r e suggested by
the diagrams of Fig. 1-37,

1- 60

a) When no command signal is applied to the gyro unit, the servo- drive
stabilizes the input axis orientation of a controlled m e m b e r on the basis of angular
deviation signals f r o m the gyro unit.
b) Command signal integration appears when an input is supplied to the gyro
unit torque motor. The resulting torque on the float causes rotation which produces
an output signal. This signal acts a s an input to the s e r v o so that the controlled
m e m b e r t u r n s in the proper direction to reduce the output signal.

Except f o r dynamic

response effects which m a y be reduced to negligible levels by proper s e r v o design,


the operation described m a y be made to rotate the controlled member about the input
axis with an angular velocity effectively proportional in magnitude to the magnitude of
the command signal.

Directional s e n s e s a r e determined by phasing o r polarity changes

of the command signal,

The resulting effect of the signal is that the angle turned

through with r e s p e c t to inertial space by the controlled m e m b e r about the input axis

is proportional to the t i m e integral of the command signal.


Another mode of gyro-unit operation that is generally s i m i l a r to the space
integrator result described above is that in which the command signals a r e applied to
the gyro unit to keep the output signal at null as the controlled m e m b e r is forced to
rotate in any a r b i t r a r y way with r e s p e c t to inertial space about the input axis.

With

this mode of operation, integration of the command signal over any given t i m e interval
r e p r e s e n t s the total angle turned through by the controlled m e m b e r during the s a m e
interval. It is to be noted that with this type of operation, the servo- drive of Fig, 1-37
has no significant function beyond providirg a shaft f o r controlled m e m b e r rotation,
In practice, command signals may be direct current, alternating current, o r
electric'al pulses. P u l s e s a r e especially suitable f o r command signals because they
m a y be e a s i l y adapted as inputs-outputs f o r digital computers, Integration with pulses
is particularly e a s y as the p r o c e s s involved is a m a t t e r of simple counting,
c) When the axis of base rotation is about an axis perpendicular to the s e r v o drive axis and the gyro unit is adjusted on the controlled m e m b e r so that i t s input axis

is at right angles to the s e r v o axis, the arrangement m a y be used to indicate the


direction of the axis about which the base is being rotated.

In effect, if the input axis

is inclined so that it r e c e i v e s a component of the base angular velocity, the gyro unit
c a s e is forced to rotate about its input axis so that it generates an output signal which
causes the s e r v o to t u r n the c a s e so that its input axis is at right angles to the base
rotation axis. By giving the gyro unit performance that enables it to respond consistently to say, one millionth of the magnitude of the base angular velocity, the
direction of the gyro unit input axis m a y be made to indicate perpendicularity to the
base rotation within an inaccuracy with the o r d e r of one a r c second.

1-6 1

This mode of operation using components now available, makes it possible to


align controlled m e m b e r s to north with sufficient accuracy for the operation of high
performance guidance systems.

1-62

d
CD

m
I

I- 6 3

STAB1LIZATION
BASE MOTION ISOLATION

GYRO UNITJ

I N P U T A X I S OF GYRO U N I T

B A S E MOTION
COMMANDED ANGULAR
V E L O C I T Y OF GYRO U N I T

SERVO SHAFT
GYRO UNIT
INPUT AXIS

ANGULAR VELOCITY OF GYRO UNIT ABOUT THE INPUT


AXIS WITH RESPECT TO INERTIAL SPACE IS PROPORTIONAL TO THE COMMAND SIGNAL ANGLE TURNED
BY GYRO UNIT REPRESENTS INTEGRAL OF COMMAND
SIGNAL

ANGULAR VELOCITY OF
BASE ABOUT AXIS
AT RIGHT A N G L E S
T O S E R V O AXIS

AXIS O F B A S E ROTATION
INPUT AXIS
OF

SERVO
GYRO UNIT
BASE

IF THE GYRO UNIT INPUT AXIS IS PLACED


AT RIGHT ANGLES TO THE SERVO SHAFT
AND THE ELECTRONIC CIRCUITS ARE ARRANGED
SO THAT THE GYRO UNIT OUTPUT SIGNAL DRIVES
THE INPUT MS TOWARD THE ORIENTATION IN WHICH
THE UNIT SENSES ZERO ANQULAR VELOCITY INPUT
THE CONFIGURATION SHOWN WILL SEEK PRPENDICIJ!LARITY
TO THE AXIS ABOUT WHICH THE EASE IS BEING ROTATED

Fig. 1-37

Basic F e a t u r e s of Illustrative Single Axis Configurations of Servo-Drives


and the Single-Degree-of-Freedom Gyro Unit t o Provide: a ) Stabilization,
b ) Space Integrators for Angular Motion, c ) Angular Velocity Axis
Direction Sensor

I- 6 4

CHAPTER I- 7
SPECIFIC FORCE RECEIVERS

All instruments designed to receive specific force depend upon a sensitive


element that is essentially an unbalanced mass. This m a s s is arranged so that it
imposes a force, o r a torque, on some m e m b e r that is r e s t r a i n e d in a calibratable
way. Figure I- 38 illustrates the u s e of a ,floated-magnetic suspension c a r r i e d
pendulum r e s t r a i n e d by a torque g e n e r a t o r fed by pulses under the control of float
angle signals. The integral of specific f o r c e (the resultant of gravity and inertia
reaction force) is obtained by counting the pulses required t o keep the float angle on
i t s null position.
F i g u r e 1-39 suggests the f e a t u r e s of a pendulous gyro- servo- drive specific
force r e c e i v e r in which the calibrated balancing torque f o r the output of the unbalanced
m a s s is provided by the gyroscopic output torque f r o m a constant speed spinning
r o t o r driven by a s e r v o to keep the output signal at its null level,

Since the required

angular velocity about the gyro input axis is proportional the balancing torque, the
angle turned by the gyro unit c a s e in a given t i m e interval is a m e a s u r e of the specific
force integral for the s a m e time.

This calibration relationship depends upon inter-

actions between two absolutely linear, non- saturable effects (both depending only on
Newton's Laws of Dynamics) so that the range of accurate operation is basically
limited only 'by s e r v o design considerations.

By using digitalizing signal g e n e r a t o r s

for the servo- axis, the integrated output appears in t e r m s of pulses suitable for direct
u s e as digital computer inputs.
The requirements of position location for center- of- mass positions to be
maintained in o r d e r to realize various drift r a t e s with typical gyro units a r e summarized in the table of Fig, 1-41,

These numbers also suggest that specific force

r e c e i v e r s must have their unbalanced m a s s c e n t e r s positioned within v e r y s m a l l


-6
tolerances if inaccuracies of 1 0 e a r t h ' s gravity a r e to be maintained o v e r a range
including some ten's of gravity.

1-6 5

I- 6 6

I- 67

SIGNAL

I INPUT

SRA

AXIS

'A3

IA
Single-degree-of-freedom gyro p a c k q e

Schematic drawing of single-degree-of-freedom gyro unit

Two-degree-of-freedom gyro package

Schematic two-degree-of-freedom gyro

PENDULOUS
MASS

SIGNAL
GENERATOR

--

"

OUTPUT
AXIS

OA

INPUT AXIS
IA

CENTER OF
M AS5 LINE

Fig. 1-40

Basic Elements of Typical Geometrical Reference Packages

I- 68

"Elx
W

9
d

L?
Fc

In

VI

al
c
W

+ E
.-

al

I- 69

CHAPTER I- 8
INERTIAL SYSTEMS

Inertial guidance s y s t e m s a l l r e q u i r e the instrumentation of reference


coordinates accurately aligned with the external space used f o r flight path r e f e r e n c t
purposes. In addition, specific force r e c e i v e r s rigidly mounted on the reference
m e m b e r a r e needed to produce signals that r e p r e s e n t specific f o r c e components,
A typical arrangement is suggested by the diagram of Fig. 1-42.
Controlled member
stabilization maintains the reference coordinates and specific force r e c e i v e r outputs
give computer inputs representing components along known axes. These outputs,
processed by the computer, give control, navigation and guidance outputs without the
necessity of geometrical transformations based on gimbal orientations.

1- 70

COMPUTING
SYSTEM

INERTIAL
REFERENCE

GIMBAL-DRIVE
POWER CONTROL

SPECIFIC-FORCE

RESOLVERS

Fig. 1-42 Basic F e a t u r e s of Inertial Guidance System with Inertial Reference


Package and Specific F o r c e Receiving Package Rigidly Mounted on a
Common Controlled Member

BIBLIOGRAPHY

1.

Schuler, M., Die Storung von Pendel-und Kreiselapparaten


durch die Beschleunigung d e s Fahrzeuges, Phys. Z.,
Vol. 24, p. 344, 1923.

2.

Ishlinsky, A. Yu., Equations for the Problem of Determining


the Position of a Moving Object by Means of Gyroscopes
and Accelerometers, Prikladnaya Matematika i Mekhanika,
Vol. XXI, NO. 6, pp. 725-739, MOSCOW, 1957.

3.

INERTIAL GUIDANCE, Pitman, G. R. (editor), John Wiley,


New York, 1962.

4.

Draper, C. S., Wrigley, W and Hovorka, J., Inertial


Guidance, Pergamon diess, New York, 1962.

5.

Markey, W. and Hovorka, J., The Mechanics of Inertial


Position and Heading Indication, Methuen, London,
1961.

6.

Broxmeyer, C., Inertial Navigation Systems, McGraw-Hill,


New York, 1964.

7.

Wrigley, W., Woodbury, R. B., and Hovorka, J., Inertial


Guidance, S. M. F. P a p e r FF-16, Inst. of the Aerospace
Sciences, New York, 1957.

8.

Draper, C. X., Wrigley, W., Hovorka, J., and Marshall, R.


E., Non-Linear Effects in Radio-Monitored Inertial
Guidance Systems, IUTAM Symposium, Kiev, 1961.

9.

Draper, C. S., Mechanization of Inertial Guidance Systems,


GYRODYNAMICS. H. Z i e d e r (editor) DX). 92-118.
Springer-Berlin, 1963.
u

10.

INERTIAL NAVIGATION ANALYSIS AND DESIGN,


O'Donnell, C. F. (editor) McGraw-Hill, New York, 1964.

11.

GUIDANCE AND CONTROL O F AEROSPACE VEHICLES,


Leondes, C.T. (editor) McGraw-Hill, New York, 1963.

12.

Fernandez, M., and Macomber, G. R., Inertial Guidance


Engineering, Prentice-Hall, Englewoocl Cliffs, 1962.

1-72

BIBLIOGRAPHY (Cont'd)

13.

Savant, C. J. , Jr., Howard, R. C., Solloway, C. B. ,


Savant C . A., Principles of Inertial Navigation, McGrawHill, New York, 1961.

15.

Cannon, R. H., Jr. , Alignment of Inertial Guidance Systems


by Gyrocompassing- Linear Theory, J. Aerospace Sci.,
Vol. 28, No. 11, 1961.

16.

Duncan, .D. B. , Combined Doppler Radar and Inertial


Navigation Systems, Proc. Nat. Electronic Conf. ,
Chicago, 1 9 58.

17.

THEORETICAL BACKGROUND OF INERTIAL NAVIGATION


FOR SUBMARINES, M. I. T. , Instrumentation Laboratory,
Report R-9, 1952.

18.

Hellman, H. , The Development of Inertial Navigation,


Navigation, Vol. 9, NO. 2, 1 9 6 2 .

19.

McClure, C. L. , Theory of Inertial Guidance, Prentice-Hall,


Englewood Cliffs, 1960.

20.

Slater, J. M. , Newtonian Navigation, Electromechanical


Design, Vol. 6, -1962.

21.

Wrigley, W. , Schuler Tuning Characteristics in Navigational


Instruments, Navigation, 1 9 50.

22.

Wrigley, W,, Single-Degree-of-Freedom Gyroscopes, M. 1. T ,


Instrumentation Laboratory, Report R- 3'1 5,
References: 1, 7-18, 24, -25, 38 and 40, 1952.

2 3.
2 4.

2 5.

GYRODYNAMICS, Ziegler, H. (editor) Springer- Berlin,


1963.
Draper, C.S., Whitaker, H.P., and Young, L.R., The
Roles of Men and Instruments in Control and Guidance
XVth International Astronautical
GROUND SUPPORT SYSTEMS FOR MISSILES AND SPACE
VEHICLES, Brown, K,, and Weiser, P. (Editors)
McGraw-Hill, New York, 1961.

1- 73

PART I1

THE NAVIGATION, GUIDANCE AND CONTROL


OF A MANNED LUNAR LANDING

David G. Hoag

DAVID G. HOAG
Associate Director, Instrumentation Laboratory
Massachusetts Institute of Technology

David G. Hoag is an Associate Director of Instrumentation Laboratory, Massachusetts Institute of Technology.

He is assigned to the technical direction o f design and

development of the guidance-navigation system the Laboratory is developing f o r the


National Aeronautics and Space Administration s Apollo spacecraft,

Prior t o his p r e s -

ent assignment, M r . Hoag was in charge of technical design of the inertial guidance s y s t e m for the Navy s P o l a r i s m i s s i l e which the Laboratory developed.
M r . Hoag was born October 11, 1925, in Boston.

He w a s graduated f r o m Chauncy

Hall School, Boston, in 1943, and entered M. I. T. that y e a r a s a m e m b e r of the Navy


V-12. He received his Bachelor of Science degree i n Electrical Engineering f r o m M. I. T.
in 1946, and his Master of Science degree i n Instrumentation f r o m M. I. T. in 1950.
M r . Hoag joined the Instrumentation Laboratory in 1946, and worked f o r s e v e r a l
y e a r s on gun fire control s y s t e m s and ship-launched m i s s i l e f i r e control s y s t e m s f o r the
Navy, He became a n Assistant Director of the Laboratory in 1955, and was placed in
charge of all technical design aspects of t h e P o l a r i s guidance system when the Laboratory

later assumed responsibility f o r that work.


196 3.

He was appointed a n Associate Director in

11-2

P a r t I1
THE NAVIGATION, GUIDANCE, AND CONTROL
OF A MANNED LUNAR LANDING
CHAPTER 11-1
THE BACKGROUND AND THE PROBLEM
O F SPACECRAFT GUIDANCE, NAVIGATION, AND CONTROL

INTRODUCTION
Among the many extensions of old disciplines and development of new technologies
needed in man' s present rush into space flight is that of the subject of this book: the
measurement and control of spacecraft position, velocity, and orientation in support of
space mission objectives. In this chapter, we w i l l introduce more specifically the nature
of the problem in order to provide a background of definition and approach for the following chapters which deal with actual details of specific problems and their solutions.
One might choose the words "spacecraft rotational and translational management"
The parameters of concern a r e the time history of
the three degrees of freedom describing spacecraft orientation and the time history of
the three degrees of freedom describing spacecraft position.
as being descriptive of the subject.

Spacecraft missions such a s those being flown today operate in phases which alter0
nate with a short period of powered or accelerated flight followed by a long period of
free-fall coasting. This is a consequence of the character of available propulsion typical in the chemical rocket. The nature of the rotational and translational management
problems differ markedly between the free-fall and thrusting accelerated conditions.
Thus it becomes convenient to separate discussions and base definitions on paired combinations of the "rotational" o r "translational" and the f'free-fall'' o r "accelerated" a s pects of the subject. This results in the following definitions of four often-used terms:
A.

"Navigation"
Translational measurement and control in free fall

B.

"Attitude Control"
Rotational measurement and control in f r e e fall

C.

"Guidance"
Translational measurement and control during acceleration

D.

"Thrust Vector Control''


Rotational measurement and control during acceleration
__

""Accelerated" here r e f e r s to that motion with respect to the free-fall arising from
non-gravitational forces.

11-3

Unfortunately two minor flaws m a r this s y m m e t r i c a l a r r a y of definitions.

First,

the p r o c e s s of "navigation" probably ought not to be constrained only to f r e e - f a l l flight.


Indeed the determination of position and velocity during any phase of flight might b e a
better definition of navigation.

We can take the view then that navigation is one of the

functions of the guidance p r o c e s s

a s w i l l be seen.

Second, using " thrust vector con-

trol" f o r t h e title associated with rotational measurement and control during accelerated
flight appears to exclude s i m i l a r operations during phases where aerodynamoc f o r c e s
not rocket thrust

a r e causing-the acceleration.

This o c c u r s during the important

phases of planetary atmospheric entry using d r a g and lift f o r c e s f o r deceleration and


steering.
Recognizing these qualifications, the following sections cover the problems of each
of the above four situations.
NAVIGATION (Translational Measurement and Control in F r e e Fall Coasting Flight)
Navigation a s defined herein is the p r o c e s s of measurement and computation to

It is
concerned only with the translational aspects of motion - i. e. position and velocity - and
is considered h e r e temporarily to be applicable only to the free- fall coasting conditions
of spacecraft. It includes those p r o c e s s e s n e c e s s a r y to determine needed trajectory
corrections a s well a s to compute the initial conditions of major powered maneuvers. In
determine the existing present position and probable future position of a vehicle.

t h i s s e n s e it has "control" aspects a s well a s " measurement" in that i t includes activity


to modify the spacecraft' s path.
In non-thrusting flight out in space, the f o r c e s on the craft which determine its
motion a r e dominated by the Newtonian gravitation attraction of the n e a r bodies
earth, moon, sun, and planets.

the

Generally the vehicle is influenced primarily by one

body and follows nearly the classical Keplerian conic path.

The effects of f o r c e s other

than that of the point m a s s c e n t r a l body can usually be treated as deviations o r perturbations to the simpler motion. A non-exhaustive listing of typical perturbing effects a r e :
(1) Mass distribution within central body, e. g. oblateness of earth, triaxiality of moon,
( 2 ) Attraction of m o r e r e m o t e bodies, ( 3 ) Atmospheric drag, (4) Solar radiation p r e s s u r e ,
(5) Meteroid impact, and ( 6 ) Magnetic and e l e c t r i c field interactions with spacecraft.

In a given situation it is usually possible to ignore a l l but a few of the perturbing


effects and predict the future trajectory of the vehicle with satisfactory accuracy many
hours to many days into the future using knowledge of present position, and velocity.
However, for a given accuracy, the prediction finally d e t e r i o r a t e s due to ignored p e r turbing effects and due to the accuracy limitations of the initial conditions and the extrapolation mod el.

I1 -4

Because of the relatively predictable nature of spacecraft t r a j e c t o r i e s in f r e e


coasting flight, continuous measurement of position and velocity is unnecessary.

Meas-

urements a r e needed periodically to c o r r e c t for the slow deviation of the actual spacecraft f r o m the predicted path.
Practical navigation measurements in f r e e coasting flight all utilize electromagnetic radiation at appropriate wavelengths to s e n s e spacial relationships among the
spacecraft and the n e a r bodies of the solar system.

These measurements can be cate-

gorized into two types: First, those made earth-based by remote tracking of the spacecraft f r o m suitable stations on the earth, and second, those made f r o m on board using
sensing devices on the craft itself.

Only the first of these h a s yet been applied; all

U. S. spacecraft and a s f a r as we know all Soviet vehicles have been navigated using
earth-based tracking measurements only.

Earth- Based Navigation Earth-based tracking f o r navigation usually uses r a d a r


frequencies with the cooperative use of transmitting beacons o r transponders on the
spacecraft being tracked.

Optical wavelengths have seen use but suffer f r o m the prob-

l e m of obscuring cloud cover,


Radio tracking for navigation is founded upon: (1) the fixed and well-known speed
of light in space, ( 2 ) the use of highly accurate time b a s e s and stable frequency sources,
and ( 3 ) the ingenuity and accuracy with which p r e c i s e phase measurement can be made
between two signals in the presence of interferring noise.
Earth- Based Range Measurement - F o r measurement of spacecraft range, the
earth station t r a n s m i t s a periodic waveform on a high frequency c a r r i e r to the spacecraft which in turn is equipped to r e - r a d i a t e this waveform back to the earth. The distance to and f r o m the spacecraft is proportional to the phase lag of the waveform a s
received f r o m the spacecraft with r e s p e c t to the transmitted waveform to the spacecraft.
Range resolution then is that fraction of a wavelength with which the phase can be m e a s -

A 100 m c c a r r i e r , for instance, h a s a wavelength of 3 m e t e r s , and range resolution


well inside this dimension is straightforward. Lower frequency modulation tones with

ured.

longer wavelengths must be used to resolve the ambiguities and thereby determine the
m o r e significant figures of the number representing measured range.

F o r spacecraft

this technique depends upon a transponder in the spacecraft which w i l l receive the t r a n s mission and r e - r a d i a t e with controlled phase-shift an appropriate correlated signal to
the ground station.
Although range tracking, a s defined, has almost micrometer resolution capability,
several limitations on the total overall accuracy exist.

The most apparent, of course,

is our knowledge of the exact speed of light. This is currently known to about 1 part i n
106 . Without calibration correction a s discussed below, this means a range e r r o r of
150 kilometers in a spacecraft distance of one astronomical unit.
reduces to the o r d e r of 400 m e t e r s .

11-5

A t lunar distances this

Range r a t e information is measured by the r a t e of change of phase shift of the


received signal, o r m o r e familiarly by the equivalent doppler frequency shift,
Earth-Based Direction Measurement - The most common direction measurement
technique f r o m earth stations is a s o r t of inverted triangulation using multiple r e c e i v e r s
on accurately known baselines.

If this baseline a r r a y is suitably short, the received

signals can be simultaneously processed in the s a m e earth-based equipment to perform


an interferometric m e a s u r e of the differences in range of the spacecraft f r o m the various
r e c e i v e r s , a s shown on Fig, 11-1. It is a technique which still offers many advantages,
particularly the fact that the spacecraft need c a r r y only a radio beacon t r a n s m i t t e r which
does not need t o be interrogated f r o m the ground.
F o r t h e s e short baselength s y s t e m s the differences in phases of the various r e ceived waveforms can b e measured with extreme precision. P 3- meter signal wavelength (100 m c ) can be resolved by phase measurement to 3 millimeters, f o r example,
utilizing techniques such a s heterodyning to a lower frequency and precision timing.

On

a 150 m e t e r baselength t h i s corresponds to 20 microradians ( 4 seconds of arc) of angular


resolution of spacecraft directions which l i e n e a r normal to the baseline.

If a space-

craft is at one astronomical unit distance, f o r instance, t h i s is position resolution of


a c r o s s the line of sight of 3000 kilometers. A t t h e s h o r t e r lunar distances this reduces
to about 8 kilometers,
For the usual horizontal a r r a y of r e c e i v e r s , it is seen that best directional a c -

curacy i s obtained for conditions with the spacecraft direction n e a r perpendicular to the
baseline. A s the vehicle gets n e a r in line with the baseline the angular resolution degene r a t e s inversely as the sine of the angle f r o m the baseline. Moreover, n e a r the horizon, earth atmospheric refraction uncertainty degenerates the total indicated direction
accuracy,

F o r g r e a t e r accuracy i n direction measurement, the baseline can be increased.


However, s e v e r a l problems i n t e r f e r e with proportional accuracy improvement of longer
baselines in comparison with the s h o r t baseline i n t e r f e r o m e t r i c s y s t e m s . First, wide
separation of the receiving stations prevents accurate, simultaneous, direct phase comparison of the received signals. Also the direction is no longer a d i r e c t function of the
range difference, a s illustrated in Fig. 11-2.

So r a t h e r than using range differences ob-

tained directly by phase comparison, the total range values f r o m the s e v e r a l stations
must be individually collected and then processed f o r determination of direction.
Realization of directional measurement accuracy improvement by increasing b a s e length r e q u i r e s that the range measurement and baseline e r r o r s accumulate l e s s rapidly
than does the baseline increase. The practicability of this can b e seen by examining the
accuracy needed in the baseline to maintain and improve the previous 20 microradian
e r r o r derived above for the short 150 m e t e r baselength. If the baselength is increased

11-6

51

B cos e

BASE L I N E , B

- R,
cose =- R , B- R2

52

A$

R,

= PHASE

DIFFERENCE ( S I , S 2 )

cose = xB &2~

11- 1

Short Base Line Interferometric Direction Measurement

R,= R ,

B2 - ~ R , B

case

11-2

Long Base Line Inverse Triangulation Direction Measurement

11- 7

to 1500 kilometers, a 2 0 microradian e r r o r r e s u l t s f r o m baseline e r r o r s of 30 m e t e r s .


To obtain improvement to 2 microradians direction e r r o r the 1500 kilometer baseline

must b e known to 3 m e t e r s .

However, at t h i s precision and better, a serious question

a r i s e s about achieving t h i s n e c e s s a r y accuracy of earth station location.


a problem of survey and geodesy.

This is clearly

The p r e c i s e knowledge of the s i z e and the shape of the

e a r t h is a question actively being pursued and about which agreement does not now exist.
Earth-based tracking ranging and directional measurements described above provide the basis f o r determining directly a l l components of the position and velocity of a
spacecraft.

The e r r o r values of o u r hypothetical models above by no means provide the

accuracy limit f r o m ground tracking.

Considerable improvement f o r a given station

a r r a y can b e demonstrated by calibration techniques in tracking t a r g e t s and applying c o r rections to fit the known target motions.
Spacecraft-Based Navigation

Spacecraft-borne navigation measurement tends

m o r e to optical frequency direction measurement r a t h e r than the radio frequency direct


ranging that is so accurate for ground tracking.

For relatively close work, however,

d i r e c t ranging using radio frequencies with rendezvous o r landing r a d a r s becomes possible, albeit n e c e s s a r y . But further f r o m the planets and other t a r g e t s , direct m e a s u r e ment of range o r range r a t e , o r the use of radio frequencies has not appeared attractive
to the designers due to the weight and power penalties.
Spacecraft onboard directional m e a s u r e m e n t s a r e those made to the n e a r bodies
t h e sun, moon, earth, and other planets.
their extreme distances,

...

The s t a r s provide no position data because of

But because of this distance they a r e most excellent r e f e r e n c e s

against which to m e a s u r e directions to the n e a r e r bodies.


In a sense, then, onboard navigation is performed by observing the n e a r bodies
relative to the background s t a r s .

This can be done indirectly by measuring the angles

sequentially from a gyro stabilized base to the s t a r s and the n e a r body.

Alternately a

d i r e c t and simultaneous measurement of t h e angle between a r e f e r e n c e s t a r and the n e a r


body with a suitable sextant-like instrument avoids an accumulation of e r r o r s with which
the f o r m e r sequential technique must cope.
The ancient sextant, updated and refined with a suitable telescope f o r image r e s o lution and with a precision angle readout of the deflecting m i r r o r , can provide in a r e a sonable s i z e an a c c u r a t e m e a s u r e of the angle between a feature of a n e a r body and a
s t a r superimposed upon that feature in the field of view.
The "feature" alluded to above is s o m e distinct point of known coordinates on the
planet to which the direction is being measured.

The center of the planetary disk natu-

rally comes to mind, but identifiable surface landmark f e a t u r e s and horizons which can
be related to planet coordinates a r e e a s i e r and m o r e accurate for visual u s e , particul a r l y under c r e s c m t illumination.

I1 -8

From sextant and sextant-like measurements, directions can be determined with


accuracies, for instance, of the order of 50 microradians to targets with an additional
target feature positional accuracy of the order of 1000 meters. F o r distances greater
than 20,000 kilometers the 50 microradians dominates. Closer to the planet, however,
navigation is limited by the location knowledge of the target features being used.
Each such angle measurement from the spacecraft provides a locus surface of
spacecraft position at the time of measurement. Several together, if made simultaneously, define position uniquely at the common loci intersection, as shown in Fig. 11-3.
In this hypothetical situation we s e e that range information is determined indirectly
from the combination of direction data in a fashion not unlike triangulation, where
the baseline is the known distance and direction between the target features of the
planets. This also, in effect, is of the same nature as stadiometric ranging, made
by measuring the apparent diameter of a planet disk.
Measurements separated in time can provide the basis for velocity determination.
To obtain three components of position in the presence of spacecraft motion, one would
desire the simultaneous measurement of at least three directional components. Practical considerations make time sequential directional measurements easier, and no
direct computation of position o r velocity is possible by purely geometric calculations,
Schemes such a s used in Apollo and described in P a r t 111of this book depend upon the use of an
onboard computer, programmed to accept the sequence of single coordinate navigation
data and the precise time each measurement occurred. Fach datum point is received
and used to update and improve in an optimum fashion the six dimensional state vector
of the spacecraft recognizing the expected e r r o r in each measurement, the current
estimate of state vector e r r o r , and the motion constraints of the spacecraft in free fall.

Ground-Based and Spacecraft-Based Navigation Measurement Comparison We can


compare the similarities and differences between navigation of a spacecraft in free fall
using earth-based tracking measurements and using vehicle-borne direction measurements a s follows:
A.

The two categories of navigation measurement complement each other in that


earth-based tracking gives strong results along the line of sight from earth,
while onboard measurement can add strength across the line of sight. The
latter is particularly accurate at distances f a r from earth and with respect
to a target planet.

B. Both categories depend upon optimum processing of data points taken over a
period of time, recognizing known measurement uncertainties and spacecraft
motion a s constrained by orbital mechanics. Both categories use the past
history of data to determine present position and velocity as limited by data
uncertainty and can predict future motion futher limited by the imperfect knowledge of the forces on the spacecraft due to the space environment.

11-9

C . Availability of earth-based navigation data f r o m a given station is dependent


t

upon the spacecraft being sufficiently above the horizon f o r that station.
Spacecraft-based navigation measurements must compete for control availability with other operations of the spacecraft.
D.

Earth-based navigation stations can support simultaneously only a limited


number of missions.

Spacecraft-based equipment, of course, is solely avail-

able for use of that mission.


E.

Earth- based navigation tracking facilities a r e most limited by economic factor:


in the attempt t o gain m o r e capability by the use of many l a r g e radio tracking
installations with complex communication networks and data processing cent e r s . Spacecraft-based navigation i s limited m o r e by the weight that can be
c a r r i e d in the s e n s o r s and data-processing computers on board.

F.

Earth- based navigation tracking facilities have the strong advantage of multiple use and r e - u s e in sequential support of many types of missions.

Space-

craft based navigation equipment is, in a sense, consumed, and only in m i s sions where the equipment is recovered would r e - u s e be possible.
G.

Earth- based navigation m e a s u r e m e n t s fail while the spacecraft is passing in


back of its target planet.

This is unfortunate since efficient orbital insertion

and t r a n s e a r t h orbital escape maneuvers always occur in back of the moon


and have a strong probability of being out of sight f o r other planets.
H.

Earth- based navigation is vulnerable to enemy action against military spacecraft.

Spacecraft-based navigation measurement can be strictly passive for

military use and is invulnerable to jamming o r sabotage.


ATTITUDE CONTROL (Rotational Measurement and Control in F r e e - F a l l Flight)

Attitude control is the p r o c e s s of aligning the spacecraft to a desired orientation


with r e s p e c t to a suitable reference framework and i n response to input commands, A s
defined h e r e the operation of attitude control applies to f r e e fall coasting flight only.

The

diverse nature of the problem is seen in t e r m s of: (1) the orientation requirements,
(b) the attitude sensing techniques, ( e ) the nature of the disturbing torques, and ( d ) the
techniques of applying the control torques, These w i l l be discussed briefly to show the
wide spectrum of problems and solutions that appear in designing attitude control systems f o r spacecraft.
The orientation requirements a r e naturally a function of the v e h i c l e ' s mission and
the associated operating constraints:

A.

Scientific payloads of a radiation o r field sensing nature generally have


pointing requirements for the sensitive axis of the instrument. Often these
aiming requirements a r e not particularly stringent, but again o t h e r s such a s

11-10

11-11

astronomical telescopes can r e q u i r e the utmost in accuracy and stability of


aiming.
B.

Spacecraft management orientation constraints generally a r e of a low o r d e r of


accuracy.

These include (1) aiming of s o l a r cells for gathering energy to

support power consuming equipment, ( 2 ) aiming of communication, telemetry ,


transponder, and beacon antennas toward earth, and ( 3 ) the maintenance of
t h e r m a l balance by controlling attitude with r e s p e c t to the sun.

C. Navigation and guidance functions r e q u i r e attitude control a r i s i n g f r o m (1) the


need to point the operating field of the navigation s e n s o r s towards the desired
portion of the sky, and ( 2 ) the need for initial pointing of the rocket t h r u s t axis
just p r i o r to ignition for a trajectory correction o r major maneuver.
This multitude of possible requirements can lead to impossible conflicting situations which a r e sometimes relieved only by mounting the lightweight instruments on
articulating gimbals t o make them a t l e a s t partially independent of spacecraft attitude.
The attitude sensing function is a l s o performed a number of ways:

A. In s o m e c a s e s radiation sensing instruments requiring pointing can be made t o


t r a c k the sensed flux themselves by providing e r r o r signals to the control

system.
B.

For e a r t h orbital spacecraft the most common attitude sensing use,s infrared

horizon detectors to indicate spacecraft orientation deviations f r o m local


vertical. These, used in conjunction with'a gyroscope reference, can a l s o
provide the attitude about t h e local vertical with r e s p e c t to the orbital plane.
This p r o c e s s is s i m i l a r to the earthbound gyrocompass in that the pendulum
is replaced by the horizon detectors and the earth' s rotation is replaced by

the rotation in orbit.


C.

Basic attitude sensing f o r small cislunar and interplanetary vehicles most


often depends upon a sun s e e k e r l t r a c k e r t o s e t up a vehicle axis with respect
to the sun, combined with a s t a r t r a c k e r offset by an adjustable angle to
acquire and t r a c k a star so as to provide attitude sensing about that sun line.

D.

Once a n orientation reference is established this can be maintained by the use


of gyroscopes to detect deviations f r o m the reference.

Gyroscopes also pro-

vide capability to m e t e r orientation changes accurately f r o m the attitude


established by other means.
The disturbance torques upset spacecraft orientation and cause the need for correction
f r o m the control system:

I1 -.12

A.

Lightweight vehicles can be affected by the relatively weak f o r c e s associated


with the space environment.

For spacecraft with l a r g e unsymmetric s u r -

f a c e s with r e s p e c t to the center of m a s s , radiation p r e s s u r e f r o m the sun is

a significant torque disturbance.


Lightweight vehicles a l s o may be affected
by interaction of electrical c u r r e n t loops o r other spacecraft magnetic s o u r c e s
with the earth s field. Electrostatic forces, unsymmetric atmospheric drag,
and the integrated effect of micrometeoroids have also been suggested a s a
source of disturbance torques,
B.

Vehicles having one long dimension resulting in a wide difference in the principal moments of inertia can be strongly affected by differential gravity f o r c e s
when n e a r a m a s s i v e planet.

C.

Spacecraft w i l l experience disturbance torques any time m a s s is thrown off.


This can occur, for instance, by the boiloff venting of cryogenic fuel o r
oxidizer o r the offloading of other wastes.

D.

Relative acceleration of m a s s e s within the vehicle cause a redistribution of


angular momentum a r i s i n g f r o m associated torques. Speed changes of onboard rotating machinery, the pumping o r sloshing of fluids, o r the p r o c e s s
of erection of solar panels o r antennas a r e examples. On manned craft the
movements of the crew cause significant disturbance.

Control torques to counteract these disturbances o r to r e - o r i e n t the vehicle can


utilize any of t h r e e phenomena.

A.

The weak f o r c e s associated with-the space environment can be utilized in a


passive o r semipassive attitude control.

Self-aligning mechanisms based

upon solar radiation p r e s s u r e , magnetic field torques, o r gravity gradient


unbalances can provide weak but often adequate restoring torques to a stable
orientation satisfactory for some missions.

Some f o r m of energy dissipation

for damping oscillations must be provided.


B.

Small reaction rocket engines a r r a y e d to provide suitable torque couples d e pend upon angular momentum t r a n s f e r to the exhausted gas.

These a r e

usually chemical o r cold gas low thrust engines designed f o r many on-off
cycles as demanded by the control loop. Control is characterized by pulsed
operation of the jet and limit cycle oscillation about the d e s i r e d attitude.
C.

Flywheel o r gyroscope momentum exchange s y s t e m s achieve control torque


by either accelerating a heavy flywheel o r precessing a spinning gyro. Unlike
the jet o r rocket s y s t e m s above, only power is consumed and operation is not
limited by the amount of working f l u i d carried.

However, t h e r e is a capacity

limit in the sense that t h e r e is a maximum momentum that can be stored by

11-13

*MOMENTS & CROSS PRODUCTS


OF INERTIA
*GAIN

*THRUST M O M E N T

*DYNAMIC C O M PENSATION
* N O I S E FILTERING

* F U E L SLOSH & M O M E N T U M LAG

*GIMBALLED
REACTION JETS

*ADAPTIVE VARIAT ION

*(AERODYNAMIC
S U RF A C E S ) *

CONTROL
SYSTEM

TORQUE

STEERING
COMMANDS

BODY

BENDING

* E N G I N E "TAIL W A G G I N G "
* ( A E R O D Y N A M I C FORCES)*
ACCELERATION

RATE C O M M A N D

TORQUE

CONTROL

VEHICLE
I

ATTITUDE CORRECTION
'

TOTAL DESIRED
ATTITUDE

FEEDBACK
SENSORS

ATTITUDE, ANGULAR RATE


ACCELERATION

RATE G Y R O S
*ATTITUDE GYROS
*ATMOSPHERIC FLIGHT
ONLY

-ORIENTATION

DY N A M ICs

GYRO STABILIZED GIMBAL SYSTEM


ACCELEROMETERS
( A N G L E OF ATTACK METER)*

11-4 Generalization of Thrust Vector Stabilization and Control

practical speeds of heavy flywheels o r gyrowheels.

Thus, in application,

these momentum exchange s y s t e m s a r e used in conjunction with periodic use


of a jet o r other type of external torques t o "desaturate" the system back
within its control range. Finally, a simple spin of the whole spacecraft itself
can often provide adequate simple means of stabilization.
The design of attitude control s y s t e m s is complicated by a number of factors.

The

classical equations of motion under assumptions of spacecraft rigidity a r e straightforward.

But even though it is theoretically possible to predict the rotational motions

of the vehicle, a simple control s y s t e m cannot make large rotational changes directly
when the desired axis of rotation does not coincide with one of the principal a x e s of
inertia.

Usually the torquing axes a r e close t o t h e s e principal axe and l a r g e rotational

maneuvers a r e made sequentially axis by axis.

This is admittedly l e s s efficient than a

hypothetical control s y s t e m that would control to the s h o r t e s t path achieved by applying

a single initial torque impulse on the n e c e s s a r y axis. This impulse would c r e a t e that
angular momentum vector which w i l l c a r r y the vehicle into the d e s i r e d orientation by
f r e e tumbling rotation where a second impulse could stop it.
The energy used in a rotational maneuver is directly a function of the speed with
which the maneuver must be accomplished. Rather than build up kinetic energy in a fast
t u r n only to cancel it again a t the destination orientation with an opposite impulse, the
designers tend towards very slow rotation r a t e s f o r the l a r g e turns when mission r e quirements permit.
Another complicating factor o c c u r s when the spacecraft c a r r i e s a significant m a s s
of fluid fuel. A practical attitude control cannot m e a s u r e the angular momentum contribution of this fluid since i t s loose coupling t o spacecraft allows it independent motion.
Again the theoretically most efficient application of control torque cannot be achieved by
simple attitude control systems.
Reaction jet control engines a r e characterized by fixed torque levels during firing
and a wearout or lifetime limit on the number or duration of individual firings.

Since

disturbance torques a r e generally less than the available control torque, the attitude
control s y s t e m provides on-off cycles of firing resulting in a limit cycle oscillation
about the desired orientation.

The total jet fuel used and the number of on-off cycles

should be minimized by optimization of the control system.

THRUST VECTOR CONTROL (Rotational Measurement and Control During


Accelerated Flight)
Thrust vector stabilization and control is the closed-loop p r o c e s s which (1) keeps
t h e vehicle attitude f r o m tumbling under the high f o r c e s of engine firing and ( 2 ) accepts
turning o r guidance steering commands to change the direction of engine-caused a c celeration.

11-15

11- 1 6

Figure 11-4 i l l u s t r a t e s a generalization of acceleration vector stabilization and


control.

In o r d e r to illustrate the variations possible, the boxes in the figure may con-

tain one o r m o r e of the a s p e c t s listed with "dot" prefix adjacent t o the boxes.

These

s y s t e m s a r e characterized by appropriate feedback t o provide a stable control of angle

or angular velocity of the thrusting vehicle. The loop also accepts input steering commands f r o m guidance to achieve a particular desired t h r u s t direction.
The design constraints on thrust vector stabilization and control s y s t e m s r e p r e sented by Fig. 11-4 v a r y considerably.

The figure lists typical variations possible in the

spacecraft body dynamics and the torque producing control devices.

Most of these c h a r -

a c t e r i s t i c s a r e not only g r o s s nonlinearities but a r e time variant and interacting a s well.


The design if f u r t h e r complicated by necessary constraints on dynamic response t o
inputs and disturbances.

It is usually r e s t r i c t e d by allowable l i m i t s on angular a c c e l e r a -

tion, angle of attack, and other variables depending on s t r u c t u r a l and controllability


considerations.

All t h i s and the u s u a l concern about reliability, weight, cost, etc.

makes design particularly difficult.


GUIDANCE (Translation Measurement and Control During Accelerated Flight)
Guidance is the p r o c e s s of measurement and computation n e c e s s a r y to provide
steering signals to the thrust vector control s y s t e m and signals to modulate engine t h r u s t
level in o r d e r to achieve vehicle acceleration to a desired trajectory. Modulation of
engine t h r u s t level in the m o r e common c a s e of a non-throttleable engine consists only
of turn-on and cutoff commands.
Earth- Based Tracking Guidance

Powered steering of some of the e a r l y USA

ballistic m i s s i l e s and of workhorse spacecraft launch vehicles used ground tracking data
in a radio command guidance illustrated in Fig. 11-5. This type of guidance is characterized by a continuous ground tracking monitor of position and velocity changes during
the powered phases and a radio command to the vehicle to change the direction of thrust
appropriately - and finally to signal t h r u s t termination. A basic requirement is a n a t titude reference system c a r r i e d aboard the vehicle.

This is illustrated in Fig. 11-5 as

the attitude feedback, implemented with g y r o s f o r instance, a s p a r t of the t h r u s t vector


control system.

Far f r o m the earth, delays occur associated with n e c e s s a r y longer smoothing of


the noisier tracking signals and delays associated with the finite speed of electromagnetic propagation.

F o r deep space spacecraft requiring s h o r t burn trajectory c o r r e c -

tions of moderate accuracy these delays a r e not significant since the ground command
need only specify the direction and length of burn required. However, for precise, long
duration maneuvers the thrust vector control alone cannot a s s u r e accuracy in metering
the direction o r magnitude of the specified velocity change. And far f r o m the e a r t h the
mentioned delays in the receipt of the steering commands make loop closure corrections

I1 . 1 7

T
I

a,

c
H
W
I

c.(

I-

11- 18

of questionable effectiveness.

Here inertial guidance is the only practical method of

powered steering control.


i

On-Board Inertial Guidance - Inertial guidance, Fig. 11-6, is based upon m e a s u r e m e n t s of vehicle motion using self-contained instruments which do not depend upon radiation sensing.

In every inertial guidance s y s t e m t h r e e types of m e a s u r e m e n t s a r e made

involving distinctive instruments: (1) angular r a t e o r direction using gyroscopic devices,


( 2 ) linear acceleration using restrained t e s t m a s s e s in a c c e l e r o m e t e r s , and ( 3 ) time

using precision r e f e r e n c e frequency sources.

The integration with time of the sensed

acceleration in the indicated direction with proper recognition of known gravity field
f o r c e s is the essence of the navigation portion of inertial guidance. The implied proce s s e s a r e accomplished in a computer with the r e s u l t of generating corrective steering
commands to the t h r u s t vector control system.

Since inertial guidance of the type

described can only integrate vehicle motion into changes in position, velocity, and orientation, accurate initial conditions a r e required in these p a r a m e t e r s before the accelerated guidance phase is started,

Initial conditions in position and velocity a r e provided

by navigation p r i o r to the accelerated phase.

Initial conditions in orientation come f r o m

the attitude control s y s t e m s o r directly f r o m s t e l l a r r e f e r e n c e s ,


One well-debated problem with inertial guidance is the presence of an increasing
e r r o r of the inertially derived orientation and navigation with time. When a n e r r o r i n
the gyro data, commonly called gyro drift, is processed in the computer the direction
of the controlled acceleration is in e r r o r . When an e r r o r in the acceleration sensing
exists, again the direction of acceleration a s well a s magnitude of acceleration and the
controlled length of motor burning a r e affected undesirably.

However, due t o the moti-

vation to perfect inertial instruments for their well adapted use in military guidance and
navigation, the technology is advanced to the point that inertial guidance performance
can be kept well ahead of needs for spacecraft missions in controlling accelerated flight.
F u r t h e r m o r e , spacecraft inertial guidance can be tolerant of e r r o r , in the s e n s e that
e r r o r s in the resulting t r a j e c t o r y usually can be measured l a t e r by navigation and c o r rected with a short burn of the propulsion.
It is perhaps pertinent to examine these last statements with r e s p e c t to two propulsion situations, which undoubtedly will exist in the future. The first is that of the
high specific impulse, low thrust e l e c t r i c engines. Here the very low thrust to m a s s
r a t i o r e q u i r e s long periods of controlled engine operation

measured in weeks.

In such

long periods, inertial guidance measurement alone without r e c o u r s e to periodic external


navigation m e a s u r e m e n t s would be unacceptable, even i f the inertial sensing w e r e p e r fect.

The inertial s y s t e m cannot s e n s e the perturbations in trajectory caused by the

imperfectly known gravitation f o r c e s .

Such s y s t e m s then require periodic navigation by

onboard or ground-tracking measurements.

It is doubtful whether these navigation checks

would be needed any m o r e often than during the coasting f r e e - f a l l phases with the m o r e
conventional chemical engine mission,
11-19

The second future propulsion situation is that which w i l l exist with high thrust
nuclear r o c k e t s providing m o r e abundant total impulse.

In this r e a l m , mission t i m e s

w i l l be shortened by longer burning to higher interplanetary velocities than permitted

with c u r r e n t chemical engines.

In spite of the l a r g e r velocity changes to be measured

during t h r u s t by the inertial sensing, the d r a m a t i c shortening of the subsequent time of


flight is enough to d e c r e a s e required measurement precision f o r the s a m e a c c u r a c y i n
a r r i v a l at the destination planet o r orbit.

CHAPTER 11-2
GUIDANCE , NAVIGATION, AND CONTROL TASKS IN THE APOLLO MISSION

Much insight into problems of space flight h a s been gained f r o m the extensive study
and hardware development during the last four y e a r s in the Apollo program for a manned
lunar landing. Using Apollo as a n example, specific spacecraft guidance, navigation,
and control t a s k s a r e illustrated i n this chapter.

Lunar Orbit InsertionIlOjj

...

Lynar Orbit
Descent
Injection(21)
-Transearth
Injection (12)

Earth Atmospheric

Translunar (7,8,9)

Earth Launch
(2,3,4)

Translunar Injection (6)

Fig. 11-7 The Overall Apollo Mission

tblf rl1 G g

3e, $4 2-.

The o v e r a l l Apollo m i s s i o n t r a j e c t o r y is summarized above. The heavy lines


correspond to the s h o r t accelerated maneuvers which a r e separated by the much longer
f r e e coasting phases.

The t r a j e c t o r y on this figure is purposefully distorted, as is

a l s o some of the following figures, in o r d e r t o show f e a t u r e s of the phases m o r e clearly,


The numbers on this figure r e l a t e to the following mission phase subdivisions:

11-2 1

APOLLO COMMAND
MODULE
SERVICE MODULE
L E M ( Inside Adaptor 1

GENERAL
I.
s 2.
3.

SATURN
INSTRUMENT
UNIT

G N & C ACTIVITY
s 1. Initial Conditions; Position,
Velocity, Mission Parameters
2. Gyro Systems Alignment

270, OOOkg

ACTIVITY
Vehicle Assembly
Checkout
Countdown

SATURN V LAUNCH VEHICLE

Fig. 11-8 Phase 1 - Prelaunch

(,.i, ,,

$,s,

7.,if .'.' I i i P
I

The prelaunch phase includes a n intensive and intricate schedule of activity to


p r e p a r e and verify the equipment f o r flight.

Automatic programmed checkout equipment

perform the exhaustive t e s t s of the m a j o r subassemblies.


During the final countdown, testing continues. Activity of i n t e r e s t h e r e concerns
the preparation of the two operating s e t s of guidance equipment for the launch.

The Sat-

u r n guidance equipment located in the Saturn Instrument Unit w i l l control the launch
vehicle.

The Apollo guidance equipment located in the Command Module (CM), where

the crew of t h r e e l i e in their protective couches w i l l provide a monitor of Saturn guidance


during launch. A third s e t of guidance equipment located in the Lunar Excursion Module
( L E M ) , which is inside the protective LEM adapter is used l a t e r n e a r the moon.

Ground support equipment communicates directly with the Saturn and Apollo CPI
guidance computers to r e a d in initial conditions and mission and t r a j e c t o r y constants a s
they v a r y a s a function of countdown status.

Both s e t s of inertial guidance s e n s o r s a r e

aligned t o a common v e r t i c a l and launch azimuth framework.

The vertical is achieved

in both c a s e s by erection loops sensing gravity, Azimuth in Saturn is measured opticallyfrom the ground and controlled by means of an adjustable p r i s m mounted on the stable
member. Azimuth i n Apollo is aligned optically on board by the astronauts and held by
gyro compassing action, During countdown, both s y s t e m s a r e tied t o a n e a r t h f r a m e
r e f e r e n c e , J u s t before liftoff, both s y s t e m s respond to signals to r e l e a s e the coordinate
f r a m e s simultaneously from the e a r t h r e f e r e n c e to the non-rotating inertial r e f e r e n c e t o
be used during boost flight.

11-22

@I

AlTlTUDE AND THRUST CONTROL


m Swivel Outer Four Engines

@I GUIDANCE

Open Loop Programmed Pitchover


to Minimfze Angle of Attack
@I NAVIGATION

Inertial Updating of Position and Veloclty


2.5 min.
-" --.
P

"

62 km

Th I

Fig. 11-9 Phase 2

E a r t h Launch First Stage

'

During f i r s t stage flight the Saturn guidance system controls the vehicle by
swiveling the outer four rocket engines. During the initial vertical flight the vehicle is
rolled f r o m its launch azimuth to the flight path azimuth. Following t h i s the Saturn
guidance controls the vehicle in an open loop pre- programmed pitch designed to p a s s
safely through the period of high aerodynamic loading. Inertial sensed acceleration
signals a r e not used during this phase to guide to desired path, but r a t h e r the l a t e r a l
a c c e l e r o m e t e r s help control the vehicle to stay within the maximum allowed angle of
attack.

Stable control is achieved in overcoming the effects of flexure bending, fuel

slosh, and aerodynamic loading by the use of properly located s e n s o r s and control networks.
Both the Saturn and Apollo Command Module guidance s y s t e m s continuously m e a s u r e vehicle motion and compute position and velocity.

In addition, the Apollo system

compares the actual motion history with that t o be expected f r o m the Saturn control
equations so a s t o generate a n e r r o r display to the crew, This and many other sensing
and display arrangements monitor the flight, If abort c r i t e r i a indicate, the crew can
f i r e the launch escape system. This is a rocket attached on a tower to the top of the
command module to lift it rapidly away f r o m the r e s t of the vehicle.

Parachutes a r e

l a t e r deployed f o r the landing.


In a normal flight the first stage is allowed to burn to near complete f u e l depletion
a s sensed by fuel level m e t e r s before f i r s t stage engine shutdown is commanded.

11-23

' I STAGE

@ A l l i T U D E AND THRUST CONTROL


Swivel Outer Four Engines

li

Ii

@I GUIDANCE- SATURN
Optimum to Desired End Conditions

NAVIGATION

Fig, 11-10 Phase 3

inertial Updating of Position and Velocity

E a r t h Launch Second Stage

Shortly after the initial f u e l settling ullage and the firing of second stage thrust,
the aerodynamic p r e s s u r e reduces to z e r o a s the vehicle p a s s e s out of the atmosphere.
At this time the launch escape system is jettisoned.

Aborts now, if necessary, would

normally b e accomplished using the Apollo Service Module propulsion to accelerate the
Command Module away f r o m the rest of the vehicle.
Since the problems of aerodynamic s t r u c t u r e loading a r e unimportant in second
stage flight, the Saturn guidance system now s t e e r s the vehicle towards the d e s i r e d
orbital insertion conditions using propellant optimizing guidance equations, Thrust
control is achieved by swiveling the outer four engines of the second stage.
During second stage flight the Apollo ,Command Module guidance system continues
to compute vehicle position and velocity.

Also this system computes any of s e v e r a l

other possible p a r a m e t e r s of the flight to be displayed to the crew for monitoring purposes. In addition, the free- fall time t o atmospheric entry and the corresponding entry
peak acceleration a r e displayed to allow the crew to judge the abort conditions existing.

11-24

i@l ATTITUDE AND THRUST CONTROL


m Gimballed Engine
Body-fixed 670N Thrusters
GUIDANCE - SATURN WITH APOLLO BACKUP
Optimum to Orbit

@L

NAVIGATION
Inertial Updating of Position a n d Velocity

Orbit Achieved
12 m i n u t e s
from Launch

Fig. 11-11 P h a s e 4

E a r t h Launch Third Stage

The third Saturn stage o r SIVB h a s a single engine for main propulsion which is
gimballed f o r thrust vector control. Roll control is achieved by use of t h e SIVB r o l l
attitude control t h r u s t e r s .
The Saturn guidance system continues to s t e e r the vehicle to orbital altitude and
speed.

When orbit is achieved, the main SIVB propulsion is shut down,

During second and third stage boost flight, the Apollo Command Module h a s the
capability, on astronaut option, to take o v e r the SIVB stage guidance function if the
Saturn guidance s y s t e m indicates failure.
sumably could be continued.

If t h i s switchover occurs, the mission p r e -

More d r a s t i c failures would r e q u i r e an a b o r t using the

In t h i s c a s e the Apollo computer is programmed t o provide


several abort t r a j e c t o r i e s : (1) immediate safe r e t u r n to earth, ( 2 ) r e t u r n to a designated
landing site, o r ( 3 ) abort into orbit for l a t e r r e t u r n t o earth.
Service Module propulsion.

SIVB engine shutdown o c c u r s about 12 minutes after liftoff a t 185 km altitude n e a r


c i r c u l a r orbit,

11-25

Roll Axis

, 185km

.Pitch
Axis

ATTITUDE CONTROL
S- IVB Thrusters

UPDATE CM GUIDANCE GYRO ALIGNMENT

NAVIGATION

DETERMINE INITIAL CONDITIONS AND AIM


FOR TRANSLUNAR INJECTION

Ground Based Tracking


On-board Landmark Sighting

Fig. 11-12 Phase 5

- E a r t h Orbit

The Apollo spacecraft configuration r e m a i n s attached t o the Saturn SIVB stage in


e a r t h orbit.

The Saturn system controls attitude by on-off commands to two of the

s m a l l fixed attitude t h r u s t e r s for pitch and to four m o r e shared f o r yaw and roll.
Ground tracking navigation data telemetered f r o m the Manned Space Flight Netw o r k (MSFN) stations is available to c o r r e c t the position and velocity of the Saturn navi-

gation system and provide navigation data for t h e Apollo navigation system,

In Apollo

the crew a l s o can make onboard navigation measurements for onboard determination of
the ephemeris by making landmark o r horizon direction sightings using a special optical
system. The Apollo inertial equipment alignment w i l l also be updated by s t a r sightings
with the s a m e optical system. F o r these measurements the crew h a s manual command
control of attitude through the Saturn system. Normally, limited r o l l maneuvers a r e r e quired t o provide optical system visability to both s t a r s and earth.
The Apollo onboard navigation m e a s u r e m e n t s include a c c e l e r o m e t e r measurement
of the s m a l l t h r u s t occurring during the p r e s s u r e venting of the cryogenic propellant

tanks of the SIVB.


Typically, the e a r t h orbital phase l a s t s for several hours before the crew signals
the Saturn system to initiate the translunar injection at the next opportunity.

11-26

Second Burn of S-IVB

- SATURN WITH APOLLO BACKUP


Steer Into "Free- Return" Trajectory
to the Moon

@I GUIDANCE

@I NAVIGATION

Inertial Update of Position and Velocity

Fig. 11-13 Phase 6

Translunar Injection

Translunar injection is performed using a second burn of the Saturn SIVB propulsion, preceeded, of course, by an ullage maneuver using the small t h r u s t e r s . Saturn
guidance and control s y s t e m s again provide the n e c e s s a r y steering and thrust vector
control to the n e a r parabolic velocity which f o r crew safety considerations puts the
vehicle on a so-called " free return" t r a j e c t o r y to the moon.

The s y s t e m a i m s to this

trajectory which ideally is constrained t o p a s s in back of the moon and r e t u r n to e a r t h


entry conditions without additional propulsion.
As before, the Apollo guidance system independently generates appropriate parame t e r s for display to the crew for monitoring purposes. It is recognized that a display of
a l a r g e e r r o r by Apollo does not necessarily indicate Saturn system malfunction because
an e r r o r in Apollo system operation could instead be the fault. The identification of the
failed system may be indicated by another of the available displays o r by ground tracking
information relayed t o the crew. If the Saturn guidance s y s t e m indicates failure, steering
takeover by the Apollo is possible without need f o r aborting the mission.
The translunar injection thrusting maneuver continues f o r slightly over 5 minutes
duration before the SIVB stage is commanded its final shutdown.

11-27

Fig. 11-14 P h a s e 7 - Transposition and Docking


The spacecraft configuration injected onto the translunar free- fall path must be r e assembled for the remaining operations.
The astronaut pilot s e p a r a t e s the Command and Service Modules (CSM) f r o m the
LEM which is housed inside the adapter in front of the SIVB stage. H e then turns around
the CSM f o r docking to the LEM. To do this the pilot has a t h r e e - a x i s left-hand t r a n s l a tion controller and a three- axis right-hand rotational controller.

Output signals f r o m

these controllers a r e processed to modulate appropriately the firing of t h e 16 low thrust


reaction control j e t s for the maneuver.

The normal response f r o m the translation con-

t r o l l e r is proportional vehicle acceleration in the indicated direction. The n o r m a l r e sponse f r o m the rotational controller is proportional vehicle angular velocity about the
indicated axis.
During the separation and turnaround maneuver of the CSM, the SIVB control syst e m holds the LEM attitude stationary. This allows for a simple docking maneuver of
the command module t o the LEM docking hatch.

The SIVB, Saturn instrument unit, and

LEM adapter a r e staged t o leave the Apollo spacecraft in the translunar configuration.
Final docking is complete l e s s than 6 . 5 hours f r o m liftoff a t the launching pad.

11-28

ATTITUDE CONTROL
16 Attitude Jets of 445NThrust each on
Service Module

Dl NAVIGATION
Ground Based Tracking and On-board
StarlLandmark Sextant Sightlngs

GUIDANCE

Used Only for Small Midcourse Corrections

Fig, 11-15 Phase 8

- Translunar

Coast

c:
/

&

/ y;'
',.
,

:.;.

54.

L;

Very soon after injection into the translunar free- fall coast phase, navigation
m e a s u r e m e n t s are made and processed to examine the acceptability of the trajectory.
These data w i l l probably indicate the need for a n early midcourse maneuver to c o r r e c t
error in the flight path before it propagates with time into l a r g e r values which would
needlessly waste correction maneuver fuel.
Once this first correction is made

perhaps a couple of hours f r o m injection

navigation activity on board can proceed a t a m o r e leisurely pace.


can be telemetered to the craft anytime it is available.

the

Ground tracking data

Using this ground data a n d / o r

onboard sextant type of landmark t o s t a r angle measurements the onboard computer can
c o r r e c t the knowledge of the spacecraft state vector - position and velocity.
The astronaut navigator can examine with the help of the computer each datum
input available

- whether f r o m ground tracking telemetered t o the craft o r taken on

- to s e e how i t could change the indicated position and velocity before he accepts
it into the computer state vector correction program. In this way the effects of mistakes
in data gathering o r transmission can be minimized.

board

11-29

@I GUIDANCE

Open Loop Steering

....Small Velocity Change

Service Module Propulsion or


Reaction Control Jets

Fig. 11-16 Phase 9

Midcourse Corrections

The navigator w i l l examine periodically the computer' s estimate of indicated uncertainty in position and velocity and the estimate of indicated velocity correction r e quired t o improve the present trajectory. If the indicated position and velocity uncertainty is suitably s m a l l and the indicated correction is l a r g e enough to be worth the effort
in making, then the crew w i l l p r e p a r e f o r the indicated midcourse correction.

Each

midcourse velocity correction w i l l f i r s t r e q u i r e initial spacecraft orientation to put the


estimated direction of the t h r u s t axis along the desired acceleration direction.

Once

t h r u s t direction is aimed, the rocket is fired under measurement and control of the guidance system.
aligned.

Use of the guidance system r e q u i r e s the inertial measurement system be

This l a t t e r is done by optical s t a r direction sightings.

Typical midcourse corrections a r e expected to be of the o r d e r of 10 m e t e r s l s e c .

If the required correction happens to be very small, it would be made by using the s m a l l
reaction control t h r u s t e r s . L a r g e r corrections would be made with a short burn of the
main s e r v i c e propulsion rocket. It is expected that about t h r e e of these midcourse velocity corrections w i l l be made on the way to the moon. The direction and magnitude of
each w i l l adjust the t r a j e c t o r y so that the moon is finally approached n e a r a desired
plane and pericynthian altitude which provides for satisfactory conditions for the lunar
orbit insertion.

11-30

/
/

/
/

ATTITUDE AND THRUST CONTROL


Pitch and Yaw; SM Glmballed Engine
Roll; SM Reaction Jets

SERViCE MODULE
PROPULSION

GUIDANCE
rn

Fig. 11-17 Phase 10

- Lunar

Steer into near Circular Orbit

Orbit Insertion

bfi[l/6-7

3 ~ 2 ~ 9
Y 0 7sc 7,

F o r lunar orbit insertion, a s with all normal thrusting maneuvers using the service propulsion of the spacecraft, the inertial guidance system is first aligned using star
sightings. Then the s y s t e m generates initial conditions and steering p a r a m e t e r s based
upon the navigation m e a s u r e of position and velocity and the requirements of the m a neuver. The guidance initiates engine turn-on, controls the direction of the acceleration
appropriately, and signals engine shut-down when the maneuver is complete.
The lunar orbit insertion maneuver is intended to put the spacecraft in a n e a r
c i r c u l a r orbit of approximately 150 km altitude.

The plane of the orbit is selected to

p a s s over the landing region on the front of the moon.

11-31

@I

NAVIGATION
Earth Based Tracking
On-board Landmark Sights
On-board Star Occultations

@I

INSPECT AND TRACK LANDING SITE

@I

CHECKOUT L E M
Align LEM Gyro System
Initial Conditions to L E M Navigation

Fig. 11-18 Phase 11 - Lunar Orbit


In lunar orbit, navigation measurements a r e made to update the knowledge of the
actual orbital motions. The navigation measurement data a r e processed in the computer
using much of the s a m e program as in the translunar phase. Several s o u r c e s of data
a r e possible.

Direction measurement to lunar landmarks o r horizons and e a r t h based

radio tracking telemetered data a r e s i m i l a r to the measurements used e a r l i e r in the


flight in e a r t h orbit, Because of the lack of lunar atmosphere, occultation time events
of identified stars by the lunar limb a r e easily made measurements,

Orbital period

measurements a r e available by timing successive p a s s a g e s over the s a m e t e r r a i n feature

or successive occultations of the same s t a r , Sufficient measurements must be made to


provide accurate initial conditions for the guidance system in the L E M for i t s controlled
descent to the lunar surface. Before separation of the LEM, this landing a r e a is examined by the crew using the magnifying optics in the command module. At this time,
direction measurements to a particular surface feature can relate a desired landing site
o r a r e a to the existing indicated orbital ephemeris in the computer. These particular
landing coordinates become p a r t of the LEM guidance system initial conditions received
f r o m the command module.
After t w o of the crew t r a n s f e r to the LEM and s e p a r a t e f r o m the Command and
Service Module (CSM), the remaining man in the CSM w i l l continue orbital navigation
a s n e c e s s a r y to keep sufficient accuracy in the indicated CSM position and velocity.

11-32

LEM Descent Engine

El GUIDANCE
Steer L E M Into Hohmann Transfer
Orbit with 15km Perllune Altitude

LUNAR EXCURSION
MODULE

The LEM guidance system w i l l have been turned on and received a checkout
e a r l i e r in lunar orbit before separation and received initial conditions f r o m the CSM.
Starting about twenty minutes before initiation of the L E M descent injection maneuver
the vehicles a r e separated, the LEM guidance system r e c e i v e s final alignment from

star sightings, and the attitude for the maneuver is assumed. The maneuver is made
using the LEM descent stage propulsion under control of the LEM guidance system,
During the short burn, the throttling capability of the descent engine is exercised as a
check of its operation. The maneuver is a 30 m e t e r p e r second velocity change to r e duce the velocity f r o m the 1600 m e t e r / s e c orbital velocity for a n e a r Hohmann t r a n s f e r
to a 15 km altitude pericynthian which is timed to occur at a range of about 370 kilome t e r s short of the final landing a r e a .

11-33

/-"

',

15Okm

CSM

El NAVIGATION
rn

@I CHECKOUT

CSM Checks L E M Orbit


rn

Fig. 11-20

Phase 13

LEM Checks Radars


LEM Checks Guidance System
Alignment

LEM Descent Orbit Coast

During the free fall phases of the LEM descent, the CSM can make tracking m e a s u r e m e n t s of t h e LEM direction for confirmation of LEM orbit with r e s p e c t to the CM.

For that p a r t of the t r a j e c t o r y in the front of the moon the e a r t h tracking can also provide a n independent check. The LEM, during appropriate p a r t s of this coasting orbit,
w i l l check the operation of its r a d a r equipment. The directional tracking and ranging
operation of the Rendezvous Radar is checked against the r a d a r transponder on the CSM.
This also provides data to the LEM computer f o r an added descent orbit check. At
lower altitudes the LEM landing r a d a r on t h e descent stage is operated f o r checks using
the moon surface return.

Alignment updating of the LEM guidance system can be p e r -

formed if desired.
The CM f r o m orbit can monitor this phase of the LEM descent using the tracking
s y s t e m s and onboard computer.

A s pericynthian is approached, the proper LEM attitude for the powered descent
phase is achieved by signals f r o m the guidance system.

11-34

El THRUST VECTOR CONTROL


3 - Axis Control Using Reaction Jets

Engine Gimballed for Trim Correction

H ENGINE
m

Near Full Throttle Under Guidance


System Control

IMI GUIDANCE
m

Inertial Sensing with Some Radar data


at Lower Altitudes

15 km

--"-350-370 km

Fig. 11-21 Phase 14

LEM Powered Descent Braking P h a s e

This phase starts a t the 15 kilometer altitude pericynthian of the descent coast
phase.

The descent engine is re-ignited, and this velocity and altitude reducing m a -

neuver is controlled by the L E M inertial guidance system,


The descent stage engine is capable of thrust level throttling over the range
n e c e s s a r y to provide initial braking and to provide controlled hover above the lunar s u r face. Engine throttle setting is commanded by the guidance s y s t e m to achieve proper
path control although the pilot can override this signal i f d e s i r e d ,
Thrust vector direction control of the descent stage is achieved by a combination
of body-fixed reaction j e t s and limited gimballing of the engine. The engine gimbal angles
follow guidance commands in a slow loop so as to cause the thrust direction t o p a s s
through the vehicle center of gravity.

This minimizes the need f o r continuous fuel

wasting torques f r o m the reaction jets,


During all phases of the descent the operations of the various s y s t e m s a r e monitored,

The mission could be aborted for a number of reasons.

If the p r i m a r y guidance

system performing the descent control is still operating satisfactorily, it would control
the abort back to rendezvous with the CSM. If the p r i m a r y guidance s y s t e m h a s failed,
a simple independent abort guidance s y s t e m can s t e e r the vehicle back to conditions for
rendezvous.
F o r a normal mission, the braking phase continues until the altitude d r o p s to about
4 kilometers o r so. Then guidance control and trajectory enter the final approach operation.

@N E N G I N E
Throttled to near ZmetersIsecIsec Under Guidance Control

0 GUIDANCE - M I X E D INERTIAL AND RADAR SENSING


w
9

llnder Automatic Control With Constant Attitude


Astronaut May Redesignate Landing Site
or
Various Levels of Manual Control
VIEW THROUGH WINDOW

Fig. 11-22 Phase 15 - LEM Powered Descent Final Approach

b/lLf/b;;'
i!
./
L

3 LS4

One significant feature of this phase is that the controlled t r a j e c t o r y is selected to


provide visability of the landing a r e a to the L E M crew. The vehicle attitude, descent
r a t e , and direction of flight a r e a l l essentially constant so that the landing point being
controlled by the guidance a p p e a r s fixed with relation to the window. A simple reticle
pattern in the window, a s shown, indicates this landing point in line with the number
indicated by computer display. The pilot may observe that the landing point being indicated is in an a r e a of unsatisfactory surface features with relation t o other a r e a s n e a r by.

He can then elect to select a new landing point f o r the computer control by turning

the vehicle about the thrust axis until the reticle i n t e r s e c t s the better a r e a .

He then hits

a " mark" button to signal the computer, r e a d s the r e t i c l e number which is in line with
this area into the computer, and then allows the guidance to r e d i r e c t the path appropriately.

This capability allows e a r l y change of landing a r e a and fuel efficient control

to the new a r e a which otherwise might have to be performed wastefully l a t e r during hover.
Automatic guidance control during the terminal phase u s e s weighted combinations
of inertial sensing and landing r a d a r data, the weighting depending upon expected uncertainties in the measurements.

The landing r a d a r include altitude measurement and a

three- beam doppler measurement of t h r e e components of LEM velocity with r e s p e c t t o


the lunar surface.

11-36

GUIDANCE

..

Various Mixtures of Manual and Automatic


Radar and Visual Correction Until Surface Obscured by Dust,
Then.. .
All Inertial For Touchdown at Low Velocity

@I TRAJECTORY
Mission Groundrules and Pilot Ootion

Fig. 11-23 Phase 16

Landing and Touchdown

At any point in the landing the pilot can elect to take over partial o r complete cont r o l of the vehicle.

F o r instance, one logical mixed mode would have altitude descent

r a t e controlled automatically by modulation of the thrust magnitude and pilot manual cont r o l of attitude f o r maneuvering horizontally.
The final approach phase w i l l end n e a r the lunar surface, and the spacecraft w i l l
enter a hover phase.

T h i s phase can have various possibilities of initial altitude and f o r -

ward velocity depending upon mission groundrules, pilot option, and computer program
yet to be decided. Descent stage fuel allowance provides for approximately two minutes
of hover before touchdown must be accomplished o r abort on the ascent stage initiated.
The crew w i l l make final selection of the landing point and maneuver to it either by tilting
the vehicle o r by operating the reaction jets f o r translation acceleration.

The inertial

system altitude and velocity computation is updated by the landing r a d a r so that as touchdown is approached good data a r e available f r o m the inertial s e n s o r s as the flying dust
and d e b r i s caused by the rocket exhaust degrade r a d a r and visual information.
down must be made with the craft n e a r vertical and at sufficiently low velocity.

11-37

Touch-

I% ACTIVITY
Checkout Systems
Exploration, Experimentation, etc

@I GUIDANCE AND NAVIGATION


Track CSM Overhead
Align Guidance for Ascent
I

Fig. 11-24 Phase 17

Lunar Surface Operations

The period on the moon w i l l naturally include considerable activity in exploration,


experimentation, and sample gatherings.

Also during this stay time, LEM spacecraft

s y s t e m s w i l l be checked and prepared for the return.

The ephemeris of the CSM in

orbit is continually updated and the information relayed t o the LEM crew and computer.
The LEM rendezvous r a d a r also can t r a c k the CSM as it p a s s e s overhead to provide
f u r t h e r data upon which to base the ascent guidance parameters.

The inertial guidance

gets final alignment f r o m optical s t a r direction sightings prior to the s t a r t of ascent.


The vertical components of this alignment could a l s o be achieved by a c c e l e r o m e t e r
sensing of lunar gravity in a vertical erection loop,
desired trajectory for rendezvous w i t h the CSM.

11-38

Liftoff must be timed to achieve the

THRUST VECTOR CONTROL


3 - Axis Control Using Reaction Jets
LEM Ascent Engine
Thrust 15,GUON

0 GUIDANCE
Inertial Sensing
Steer to Rendezvous Transfer Orbit

Fig. 11- 25 P h a s e 18

LEM Ascent

t/tr!c,.$'.

3c;3 L b

Normal d i r e c t ascent launches a r e timed and controlled to cutoff conditions r e sulting in a coasting intercept with the CSM. Emergency launches f r o m the lunar s u r face can be initiated at any time by entering a holding orbit a t low altitude until the
phasing is proper f o r t r a n s f e r to the CSM.

A desirable constraint on all ascent powered

maneuvers a s well a s abort maneuvers during the landing is that the following coasting
trajectory be n e a r enough circular so a s to be c l e a r of intersection with the lunar s u r f a c e , This is a safety consideration to allow f o r the possibility of failure of t h e engine
to r e s t a r t .

If the LEM engine thus fails, the LEM can then safely coast until a pickup

maneuver by the CSM is accomplished.


The initial p a r t of the ascent trajectory is a vertical r i s e followed by pitchover a s
commanded by the guidance equations.

The ascent engine maneuvers a r e under the con-

t r o l of the LEM inertial guidance system.

The engine has a fixed mounted nozzle.

T h r u s t vector control is achieved by operation of the sixteen reaction j e t s which a r e


mounted on the ascent stage, The engine thrust cannot be throttled but the n e c e s s a r y
signals f r o m guidance w i l l terminate burning when a suitable rendezvous coast trajectory

is achieved.

11-39

0 NAVIGATION
Rendezvous Radar Data Used to
Determine Velocity Corrections

VELOCITY CORRECTIONS
Using Inertial Guidance; As Many
as Needed to Achieve I ntercepi
with CSM

".
L
"
"

Fig. 11-26 Phase 19 - Midcourse Rendezvous


If the launch point l i e s in the plane of the CSM orbit, efficient ascending coasting
t r a j e c t o r i e s would cover 180 degrees central angle to the rendezvous point.

Several

effects w i l l cause the launch point to be removed f r o m the CSM plane resulting in t r a jectories either somewhat m o r e o r somewhat less than 180 degrees.
Immediately a f t e r injection into the ascending coasting rendezvous trajectory, the
rendezvous r a d a r on the LEM w i l l s t a r t making direction and range measurements to the
CSM upon which the LEM computer w i l l b a s e its navigation using a p r o c e s s almost
identical t o that used in navigation of the midcourse phase between e a r t h and moon.
F r o m this navigation the LEM computer w i l l determine s m a l l velocity corrections to be
made by LEM reaction control j e t s to establish the collision o r intercept trajectory with
the CSM m o r e accurately.

These corrections will be made as often a s the r a d a r based

navigation measurements justify.

The coasting continues until the range to the CSM is

reduced t o approximately 10 kilometers when the terminal rendezvous phase begins.

11-40

r*
MEASUREMENT

Rendezvous Radar

I4 MANEUVER
Several Thrusts to Reduce Relative Velocity
to zero at Point Near CSM

_"

DOCKING
Under Manual Control

TYPICAL MANEUVER
H I STORY

Fig. 11-27 Phase 2 0

Terminal Rendezvous and Docking

The t e r m i n a l rendezvous phase consists of a s e r i e s of braking t h r u s t maneuvers


Y

under control of the LEM guidance system which u s e s data f r o m its inertial s e n s o r s and
the rendezvous r a d a r . The objective of these operations is to reduce the velocity of the
LEM relative to the CSM to z e r o a t a point n e a r the CSM.

This leaves the pilot in the

LEM in a position to initiate a manual docking with the CSM using the translation and
rotation control of the LEM reaction jets.
Although these maneuvers would normally be done with the LEM, propulsion o r
control problems in the LEM could r e q u i r e the CSM to take the active role.
After final docking the LEM crew t r a n s f e r to the CSM and the LEM is then jettisoned
and abandoned in lunar orbit.

>a

I1-4 1

Midcourse

10.3m

@I NAVIGATION
Earth and Spacecraft Based Measurements
VELOCITY CORRECTIONS

To Achieve Accurate Earth Entry

Fig. 11-29 Phase 22

T r a n s e a r t h Coast

The t r a n s e a r t h coast is very s i m i l a r t o the translunar coast phase.

During the

long coasting phases going t o and f r o m t h e moon the s y s t e m s and crew must control the
spacecraft orientation a s required. Typical midcourse orientation constraints a r e those
t o a s s u r e the high gain communication antenna is within its gimbal limits to point to e a r t h
o r that the spacecraft attitude is not held fixed t o the local heating effect of the sun f o r
too long a period.
During the long periods of free- fall flight going to and f r o m the moon when the
inertial measurement system is not being used f o r controlling velocity corrections, the
inertial system is turned off to conserve power supply energy.
Onboard and ground-based m e a s u r e m e n t s provide f o r navigation upon which is based

a s e r i e s - normally t h r e e - of midcourse correction maneuvers during t r a n s e a r t h flight.


The a i m point of these corrections is the center of the safe e a r t h entry c o r r i d o r suitable
f o r the desired landing area. This safe c o r r i d o r is expressed as a variation of approximately * 32 kilometers in the vacuum perigee. A too-high entry could lead to a n uncontrolled skipout of the atmosphere; a too-low entry might lead to atmospheric d r a g a c celerations exceeding the crew tolerance,
After the final safe entry conditions a r e confirmed by the navigation before entry
phase starts, the inertial guidance is aligned, the Service Module is jettisoned, and the
initial entry attitude of the Command Module is achieved.

11-43

.
.
.

@i ACCELERATION VECTOR CONTROL


By "RolI"about wind axis to direct
l i f t vector up, down, left, or right
STABILIZATION
PITCH &YAW jet operation to damp
oscillation about trim orientation

JETS

GUIDANCE

.Steer to safe entry


Steer to landing recovery area

-V

Fig. 11-30

Phase 23

Earth Atmospheric Entry

b'.

,'

42 G. '7.7

Initial control of entry attitude is achieved by guidance system commands to the


12 reaction jets on the command module surface, A s the atmosphere is entered, aerodynamic forces create torques determined by the shape and center of m a s s location. If
initial orientation was correct, these torques a r e in a direction towards a stable t r i m
orientation with heat shield forward and flight path nearly parallel to one edge of the
conical surface. The control system now operates the reaction jets to damp out oscillation about this t r i m orientation. The resulting angleof attack of the entry shape causes
an aerodynami'c lift force which can be used for entry path control by rolling the vehicle
about the wind axis under control of the guidance system. Range control is achieved by
rolling so that an appropriate component of that lift is either up o r down as required,
Track o r across range control is achieved by alternately choosing as required the side
the horizontal lift component appears.
The early part of the entry guidance is concerned with the safe reduction of the high
velocity through the energy dissipation effect of the drag forces. Later a t lower velocity
the objective of controlling to the earth recovery landing area is included in the guidance.
This continues until velocity is reduced and position achieved for deployment of a drag
parachute. Final letdown is normally by three parachutes to a water landing.

11-44

CHAPTER 11-3
GUIDANCE , NAVIGATION , AND CONTROL INSTRUMENTATION IN APOLLO

The choice of sensors and data processors for guidance, navigation, and control
used in Apollo is governed by the nature of the spacecraft and its mission a s described
in the previous chapter. Two Apollo design guidelines w i l l be mentioned at the outset.
First, although full use w i l l be made of all earth-based help, the spacecraft sys-

tems a r e designed to have the capability of completing the mission and returning without
the use of earth-based tracking data o r computation support. This provides protection
against critical lack of earth coverage o r failure in communication. However, earthbased data w i l l be available most of the time which w i l l be supported by measurements
from the onboard equipment.
The second guideline recognizes the diverse nature of the mission and the variations in spacecraft configuration. The guidance, navigation, and control equipment is
designed to provide a great deal of flexibility in its utilization. This is manifest in the
fact that identical subsystems a r e used in the two independent systems controlling the
command module and the lunar excursion module. This flexibility extends also to the
development of the necessary operation equations expressed in the flight computer programs. The unified approach of these to handle the various thrusting and coasting computation chores with a universal compact set of programs is described in P a r t 111.
In this chapter w e w i l l describe the selection and design of hardware.
INERTIAL MEASUR-EMENT SYSTEM
The choice of inertial guidance over radio command guidance can be easily
justified . . perhaps most dramatically by recognition of the velocity change maneuvers
which necessarily must occur in back of the moon. Here the guidance measurements
must be made by onboard sensors during the lunar orbit insertion and escape maneuvers
out of sight of the earth where ground data a r e not available. Even were it not f o r the
fact that the earth is blind with respect to these maneuvers, it is extremely doubtful
radio command could function for large velocity change maneuvers at lunar distances.

The choice of inertial guidance mechanization might not be so obvious. The two
major configurations f o r inertial measurements a r e : (1) gyro stabilized gimbal-mounted
platform and ( 2 ) vehicle frame mounted sensors. Each has advantages.

11-45

The gyro stabilized gimballed platform h a s had many y e a r s of s u c c e s s and experienced gained primarily by i t s use in guidance of military ballistic m i s s i l e s . It
clearly has most superior performance due, in l a r g e part, because the gyros and a c c e l e r o m e t e r s a r e kept non-rotating by the isolation provided by the gimbals and t h e i r
servos.

Finally, the outputs a r e in a convenient form.

appear directly a s the angles of the gimbals.

Vehicle attitude Euler angles

Acceleration measurement appears di-

rectly a s components in the non-rotating coordinate f r a m e of the stable m e m b e r


"platform".
Alternately, the vehicle f r a m e o r body-mounted inertial s e n s o r s offer promise of
dramatic savings i n size, weight, and convenience in mounting. Unlike the gyros on the
gimballed system which m e r e l y must indicate the s m a l l deviations f r o m initial attitude
for closed loop gimbal control, the body-mounted gyros must m e a s u r e precisely the
whole angular velocity experienced by the vehicle.

Moreover, problems are introduced

in achieving good gyro and a c c e l e r o m e t e r performance because of this l a r g e angular


velocity the units must tolerate about all axes.

Finally the outputs a r e not always in a

d i r e c t useful f o r m , Angular orientation of the vehicle is indicated only by properly


transforming and integrating the body-fixed coordinates of angular motion indicated by
the gyros into either an E u l e r angle s e t o r a m a t r i x of direction cosines,

With either of

these, the body-mounted a c c e l e r o m e t e r signals can be resolved f r o m the rotating space-

craft coordinates into an inertial f r a m e . All these calculations r e q u i r e a computer of


considerable speed and accuracy t o prevent accumulation of excessive e r r o r .
The choice made in Apollo f o r both the Command Module and LEM spacecrafts was
the use of the gimbal stabilized m e m b e r mounting of the s e n s o r s f o r the p r i m a r y s y s tems.

The superior demonstrated performance provides a conservative margin of safety

in economical use of rocket fuel f o r the m a j o r mission completion maneuvers. The


secondary backup o r abort guidance s y s t e m s in each spacecraft, however, capitalize
upon the s i z e and convenient installation advantages of body-mounted s e n s o r s ,

Here the

m o r e modest performance is quite ample f o r the crew safety abort maneuvers in c a s e of


p r i m a r y guidance system failure.
The Apollo p r i m a r y guidance gimbal system

o r IMU f o r Inertial Measurement

is shown schematically i n Fig. 11-31. This IMU is seen to c a r r y t h r e e singledegree-of- freedom gyros which provide n e c e s s a r y e r r o r signals to stabilize in space the

Unit
-

orientation of the inner m e m b e r by servo d r i v e s on each axis.


rotational a x e s of the gimbal s y s t e m a s shown in the figure.

There a r e t h r e e of these

A t h r e e degree of gimbal

system such a s this can present problems due to a phenomena called "gimbal lock".
Gimbal lock would occur when the outer axis is c a r r i e d by spacecraft motion to be
parallel to the inner axis.

In t h i s position, a l l t h r e e axes of gimbal freedom lie in a

plane and no a x i s is in a direction to absorb instantaneously rotation about an axis p e r pendicular to this plane.

Thus, at gimbal lock the inner stable m e m b e r can be pulled

11-47

..

k
0

Lcr

cv

M
I

H
H

11-48

off of its space alignment. Even though a three-degree-of-freedom gimbal system


allows geometrically any relative orientation, the required outer gimbal angular acceleration needed a t gimbal lock to maintain stabilization w i l l exceed servo capability.
One direct solution to gimbal lock problems is to add a fourth gimbal and axis of
freedom which can be driven so a s to keep the other three axes from getting near a
common plane. However, the cost in complexity and weight f o r a fourth gimbal is considerable. Fortunately, in Apollo the operations with the IMU are such that gimbal lock
can be easily avoided, and a simple three-degree-of-freedom gimbal system is entirely
satisfactory. This w i l l be made clear in the following paragraph.
The Apollo IMU w i l l normally be turned off during a l l long coasting periods not
requiring its use. This is done primarily to save power and corresponding fuel cell
battery reactant. (Reactant savings o f the order of 20 kilograms have been estimated. )
F o r this reason, the guidance system provides for inflight inertial system alignment
against s t a r references before the start of each accelerated phase of the mission. This
allows the inner stable member alignment to be chosen for each use in the most logical
orientation. Simplifications can result in the computer generation of steering commands
if the "X" accelerometer axis on the stable member is aligned in some direction near
parallel to the expected thrust (or entry atmospheric drag). This happens also to be
optimum with respect to inertial sensor measurement e r r o r effects in velocity measurement. Since the X accelerometer is perpendicular t o the inner gimbal axis, the direction of this inner axis can be chosen a s required. F o r each mission phase involving
rocket burning o r atmospheric drag, the trajectory and the thrust o r drag lie fairly
close to some fixed plane. The inner gimbal axis is then aligned somewhere nearly
perpendicular to this plane. A l l required large maneuvers result mostly in inner gimbal motion, thus avoiding the difficulty of approaching gimbal lock associated with large
middle gimbal angles. Finally, because large roll maneuvers a r e desirable (for instance during entry for the Command Module) the outer gimbal axis is mounted to the
spacecraft along o r near the roll axis so that no restriction on roll maneuver ever exists.
Because the details of the design and operation of the critical inertial sensors gyros and accelerometers - to be mounted on the stable member of the gimbal system
a r e of particular importance and interest, this subject is covered separately in Part IV.
However, an overall view of the inertial measurement unit is shown in Fig. 11-32. In
this photograph the spherical gimbal halves and case cover a r e removed to show the appearance of the components mounted on the stable member and on the axes of the gimbals.
INERTIAL SYSTEM ALIGNMENT
A s mentioned above, the inertial measurement system is turned off during the

longer free-fall coasting periods to conserve power supply energy. Even were it not
for this, unavoidable drift of the inertially derived attitude reference would require

I1 -49

cl

11-50

periodic in-flight alignment to the p r e c i s e orientation required f o r measuring the l a r g e


guided maneuvers.

The u s e of identified s t a r directions for the inertial system align-

ment introduces the question of physically relating the sensed s t a r direction to inertial
system stable member orientation.

The problem f r o m one point of view could be mini-

mized by mounting the s t a r s e n s o r o r s e n s o r s directly on the stable m e m b e r itself.

This

would impose a most s e v e r e limitation of field of view of sky available and puts unpermissible constraints on spacecraft attitudes during the alignment,

Even a measured

two degree of rotational freedom of the s t a r s e n s o r axis on the stable member limits
flexibility and compromises design m o r e than can be tolerated.
The alternative of mounting the s t a r s e n s o r telescope separately n e a r the spacecraft skin where its line of sight can be articulated to cover a l a r g e portion of the sky
means f a r m o r e freedom in spacecraft attitude during inertial system alignment. In
Apollo a rigid s t r u c t u r e called the navigation b a s e which is s t r a i n - f r e e mounted to the
spacecraft provides a common mounting s t r u c t u r e for the s t a r alignment telescope and
the base of inertial measurement gimbal system. Figure 11-33 shows this arrangement
for the Command Module system.

(In this photograph the eyepieces of the optics a r e

not attached. ) By means of precision angle t r a n s d u c e r s on each of the axes of the telescope and on each of the axes of the inertial s y s t e m gimbals, the indicated angles can
be processed in the onboard computer to generate the s t a r direction components in
inertial system stable member coordinates.

This provides the computer with p a r t of

the needed stable member orientation data, except no information is provided f o r rotation about the s t a r line. The use of a second s t a r , a t a n angle far enough removed f r o m
in line with the f i r s t , completes the f u l l t h r e e - a x i s stable member orientation m e a s u r e ment. With this information the stable m e m b e r orientation can then be changed under
computer command, if desired, to the orientation optimum for u s e of the guidance
maneuvers.
It is recognized that the above procedure has many s o u r c e s of e r r o r in achieving
inertial system alignment.

F o r example, each axis of rotation of the s t a r telescope and

the inertial s y s t e m gimbals must be accurately orthogonal ( o r a t a known angle) with


r e s p e c t to the adjacent axis on the s a m e structure.

This is a problem of precision

machining, accurate bearings, and stable s t r u c t u r e s .

Each angle transducer on each

axis of the star telescope and the inertial system gimbals must have minimum e r r o r in
indicated angle. This includes initial zeroing, transducer angle function e r r o r s , and
digital quantization e r r o r s for the computer inputs. By careful attention to minimizing
each of these and other e r r o r sources, probable Apollo inertial system alignment e r r o r
of the o r d e r of 0 . 1 milliradian is achieved, an accuracy which exceeds requirements by

a comfortable margin.

11-51

LANDM

SIGHT

TRUN
DRI

BEAM SPLITTER
A N D POLARIZE

SHAFT
DRIVE

ADJUSTABLE
LANDMARK
EYEPIECE

I MAGNIFICATION 28 x

11-34

Sextant Schematic

11-35

81 Scanning Telescope Schematic

OPTICAL MEASUREMENT SYSTEM


Besides providing for inertial system alignment a s described above, the optical
system also provides the onboard measurement capability for orbital and midcourse
navigation of the command module. The single-line-of-sight direction measurement
referenced to the stable member used for inertial system alignment can be well utilized
also in low earth or lunar orbit for navigation. However, for onboard navigation during
the translunar and transearth phases, accuracy requirements a r e met only by a twoline-of - sight sextant type of instrument.
Two separate Command Module optical instruments a r e mounted on the navigation
base which also supports the inertial measurement unit. These a r e the two-line-of-sight
sextant and the single-line-of-sight scanning telescope.

The sextant and its features a r e illustrated diagramatically in Fig. 11-34.. It is


essentially a two-line-of-sight instrument providing magnification for manual visual use
as w e l l as special sensors for automatic use. It is seen in the figure that one of the lines
of sight of the sextant, identified with the landmark side of the navigation angle, is undeflected by the instrument and is thereby fixed to the spacecraft. To aim this line,
then, the spacecraft must be turned in space appropriately by means of orientation
commands to the attitude control system. The second l i n e identified with the star side
of the navigation angle can be pointed in space through the use of two servo motor drives
illustrated schematically. One axis of this motion - called the shaft axis - is parallel
to the landmark line and changes the plane in which the navigation angle is measured by
rotating the head of the instrument as a whole. The second axis - trunnion axis - sets
the navigation angle by tilting of the trunnion axis m i r r o r , A precision angle data transducer on this m i r r o r provides a direct measure of the navigation angle for the navigation
routine of the computer. An angle transducer on the shaft axis completes the data needed
by the computer of the indicated s t a r direction when the instrument is used for inertial
system alignment.
The light arriving along the landmark is polarized before being combined in the
beam splitter mirror with the light along the s t a r line so that the navigator can adjust
the landmark background brightness relative to the star intensity by means of an eye,piece polarizer. The sextant also uses the trunnion m i r r o r in conjunction with a star
tracker sensor to provide automatic star tracking e r r o r signals to the shaft and trunnion
drives.

'

Mounted with its sensitive axis along the landmark line is a second automatic detector called a horizon photometer. This device senses the brightness of a small portion of the sun illuminated horizon for use a s one side of the navigation angle. This is
described in more detail in P a r t V.

11-53

11-36

11- 37

Telescope View

Midcourse Navigation

Spacecraft Orientation - Midcourse Navigation Sighting

I1- 54

Because the 28 power magnification of the visual section of the sextant results in
less than a 2 degree diameter field of view, the second instrument, the scanning telescope, provides a wide field acquisition capability for the sextant to find and acquire
objects in the sky. The u s e of an entirely separate optical instrument rather than a
combined variable power instrument using one set of line-of-sight articulation drives is
justified by the simpler mechanical and optical configuration and the sighting redundancy
two units provide.
The scanning telescope illustrated in Fig. 11-35 has shaft and trunnion pointing of
its single line of sight. The shaft angle always is made to follow the sextant shaft angle
by servo action when the optics system power is on. The trunnion can be selected by the
astronaut to (1) follow the sextant trunnion and hence look along the star line, ( 2 ) be driven
to zero and hence look along the landmark line, o r (3) be driven to a fixed angle of 25
degrees. This latter provides for ease in simultaneous acquisition of landmark and s t a r
since the scanning telescope w i l l indicate the image along the landmark line by a reticle
point 25 degrees from the center of the field and w i l l indicate possible s t a r s available by
trunnion motion in the sextant field of view on a diametrical reticle line. Figure 11-36
shows the view through the scanning telescope during acquisition, Generally, the navigator w i l l preset the trunnion to the expected navigation angle indicated by the computer
a s the preliminary step in the acquisition process.
Under manual visual control, the shaft and trunnion drives of the telescope a r e
commanded by a left-hand two-axis controller. By this controller, the navigator can
point the scanning telescope and the sextant s t a r line. At his right hand, the navigator
has spacecraft attitude controllers with which he can rotate the spacecraft to position
the landmark line. His midcourse cislunar sighting strategy is to set up the measurement situation illustrated in Fig. 11-37'. With his right hand controls, he gets the
identified landmark within the field of view of the sextant a t a slow spacecraft rotation
drift. He need then only provide occasional minimum impulses from the appropriate
attitude jets to keep the landmark within the field while with his left hand he positions
the s t a r image to superimposition on the landmark. When this is achieved, Fig, 11-38,
he pushes a "mark" button which signals the computer to record the navigation trunnion
angle and time. From these data the computer updates the navigation state vector.
The unity power wide field of view of the scanning telescope is also suitable for
navigation direction measurements to landmarks in low earth o r moon orbit. The wide
field of view makes landmark recognition easy. Landmark direction measurement a c curacy of the order of 1 milliradian as referenced to the pre-aligned attitude of inertial
system and as limited by the unity magnification is sufficient for landmark ranges under
a few hundred kilometers.
In low orbit, the inertial measurement gimbal system must be on and pre-aligned
with two s t a r sightings. Then the navigator acquires and tracks landmarks a s they

11-55

pass beneath him, pushing the mark button when he judges he is best on target, Fig. 11-39.
The computer then records optics angles, inertial measurement unit gimbal angles, and
time to provide the navigation data.
In all uses of s t a r s and landmarks for navigation the computer must be told by the
navigator the identifying code o r coordinates of the star andlor landmark. These appear
on the navigator' s maps and charts to help his memory.

ONBOARD COMPUTER
The relatively large amount of onboard data processing required for Apollo guidance, navigation, and control can be met only by the capabilities of specially designed
digi'tal computer. The special requirements define a computer which would provide for:
1. Logic, memory, word length and speed capability to f i t the needs of the

problems handled.
2.

Real time data processing of several problems simultaneously on a priority


basis.

3.

Efficient and yet easily understood communication with the astronauts for
display of operations and data a s well a s manual input provisions for instructions and data.

4.

Capability of ground control through radio links a s w e l l a s telemetering of


onboard operations and data to the ground.

5.

Multiple signal interfaces of both a discrete and continuously variable nature.

The design features of this computer a r e covered in detail in Part VI. But perhaps the many input and output signals should be discussed briefly here because of the
large part these interfaces make in determining the system configuration and in understanding system tasks and operation. Rather than a listing of interfaces the important
ones w i l l be discussed by groups in the following paragraphs.
Inputs to the computer of a discrete o r two-state nature a r e handled a s contact
closures o r voltage signals. These offer no difficulty except for the computer activity
needed to keep appraised of them. Important urgent signals of this nature - such as an
abort command o r the time critical "mark" signals go to special circuits which interrupt computer activity to be processed before other activity is resumed o r modified,
Less critical signals indicating states of the various equipment or requiring l e s s urgent
action a r e examined by the program periodically a s necessary.

11-38

Sextant View

Midcourse Navigation

11-39

Telescope View

11- 57

Orbital Navigation

II- 40

Ball Attitude Indicator

II- 58

Discrete signal outputs a r e of two types. Time critical ones such a s that which
signals engine thrust cutoff consist of high frequency pulse trains which a r e gated on at
the time of the programmed event and detected remotely where the action is requested.
Slower discrete outputs a r e either gated d-c voltages o r relay contact closures set by a
state matrix w,hich drives the appropriate relay coils. The majority of these relays a r e
used to set the states of the electroluminescent number display readout of the computer
display and keyboard. Others change operating modes of the associated spacecraft systems o r a r e used to light status o r warning lights.
Direct earth communications to and from the computer requires circuits associated
with the interface with the radio receiver and transmitter to convert between the serial
code of the telemetry and the parallel format of the computer.
Other variables into the computer a r e handled by input counters which sum pulses
transmitted as the indicated variable changes through fixed increments. Velocity increments, for instance, measured by the inertial system accelerometers a r e handled
in this way.
Some variable outputs, such as the command torquing of the inertial system gyros
to change alignment, appear a s output increment pulses on appropriate lines.
Perhaps the most difficult class of computer interfaces is handled by the use of
auxiliary pieces of equipment called Coupling Data Units - o r CDUs for short. The CDUs
provide the means for coupling with the digital computer the sine and cosine analog signals from the resolver type of angle transducers used on the optics and inertial gimbal
system axes, There a r e five of these CDUs, one each associated with optics shaft,
optics trunnion, and the three axes of the inertial unit gimbal system. The details of
the operation and construction of the CDUs a r e described in P a r t IV.
DISPLAYS AND CONTROLS
The preceeding sections have introduced the needs and characteristics of three
Apollo guidance, navigation, and control subsystems: (1) the inertial measuring equipment, ( 2 ) the optical measurement equipment, and ( 3 ) the digital computer data processing equipment. Because Apollo is by its very purpose a manned mission, the provision for system operation by the crew is essential. This identifies the need for the
fourth subsystem: the displays and controls.
The provisions to involve the astronaut might appear as an unnecessary complication. Indeed many tasks a r e best left to the machine: those that a r e too tedious o r
require too much energy, speed of response, o r accuracy outside m a n ' s capabilities,
But the utilization of man in many of the tasks of guidance, navigation, and control more
than pays for the display and control hardware needed. His involvement without any
doubt enhances mission success significantly. Consider man' s judgement and

11-59

adaptability, his decision-making capability in the face of the unanticipated, and his
unique ability to recognize and evaluate patterns. Of this latter, consider his unsurpassed faculty to pick out a particular navigation star from the heavens o r to evaluate a
suitable touchdown spot on the moon.
Displays and controls were designed in Apollo to provide the crew with visability
into and command over the guidance, navigation, and control tasks. In most of these
tasks, then, the astronaut can select either to be intimately involved in the procedures
o r allow f u l l automatic operation which he w i l l be able to monitor at his discretion.
In the command module, the navigator has displays and controls illustrated in
Fig. 11-41. The eyepieces of the sextant and scanning telescope appear prominently beside each other. Just below these eyepieces is a control panel used primarily for operating the optics. The left-hand optics hand controller and the right-hand spacecraft
attitude minimum impulse controller appear at the top of this panel. A t the bottom of
this panel a r e operating mode selector switches. The inertial system mode controls
and displays appear to the left and above the eyepieces. Directly to the left a r e the
five coupling data units with a display of the associated variable of each. To the right
is the numerical readout and keyboard associated with the computer. Features of this
a r e described with the computer in Part VI.
Tabulations of data, lists of procedures, and maps and charts of landmarks and stars
w i l l be provided in a bound book o r could alternately be projected on a microfilm projector. Space for this projector appears near the top of Fig. 11-41. Controls to operate the
film drives and projection lamp a r e seen on the center panel.
The numbered modules below the optics control panel contain the miscellaneous
analog electronics that operate the equipment. And below this is the digital computer.
Separated from the displays and controls described above and on the main panel
in front of the pilot s couch certain important guidance data a r e displayed, A second
display and keyboard of the digital computer is mounted here. This unit is functionally
in parallel with the one used by the navigator so that the majority of guidance and navigation functions can be operated and observed from either station.
Also visible to the pilot is a ball attitude indicator and associated needles, Fig.
11-40. The spacecraft orientation is indicated to the pilot by the attitude of the ball which
is driven in three axes by the three axes of the inertial measurement unit gimbals. Also,
the three components of attitude e r r o r generated by the guidance system a r e displayed by
the position of three pointers which cross the face of the instrument. Vehicle attitude
rates measured by three vehicle mounted rate gyros a r e displayed by three more pointers
around the sides of the instrument.
Other displays showing guidance, navigation, and control system status also a r e available f o r the pilot on the main panel along with a complex a r r a y of equipment associated
with other systems for controlmode selection, and display.
11-60

11-41

Displays & Controls - Command Module Lower Equipment Bay

11- 61

EQUIPMENT INSTALLATION IN SPACECRAFT


The installation of the guidance and navigation equipment in the command module
is illustrated in the cutaway view, Fig. 11-42.

This sh3ws the navigator operating the

displays and controls a t the lower equipment bay where the majority of t h e guidance and
navigation equipment is located.
spacecraft.

Other control equipment is distributed around the

During launch boost into e a r t h orbit and during r e t u r n entry when the a c -

celeration f o r c e s a r e high, the navigator must leave his station a s shown and l i e in the
protective couch in the center between his companions.

Sufficient controls and displays

a s described above a r e on the main panel in front of the couches t o perform all the guidance and navigation functions except f o r those requiring visual use of the optics. Thus,
f o r the limited period near the earth when the navigator cannot be stationed in front of

the eyepieces at the lower equipment bay, use of optics is through the automatic fea-

t u r e s only.
The installation in the LEM is shown in Fig. 11-43.

The inertial measurement unit

(IMU), the LEM guidance computer (LGC), the coupling data units (CDUs), and support
electronics of the power s e r v o assembly (PSA), a r e all identical to those used in the
command module. Since the LEM activity, when separated f r o m the command module,
does not r e q u i r e optical navigation sightings, a simpler optical alignment telescope i s
installed on a navigation b a s e with the inertial measurement unit and i s used only f o r
aligning the stable m e m b e r of the latter.

Also unique to the LEM a r e the two r a d a r s .

The rendezvous r a d a r is mounted n e a r the inertial unit so that direction data can be
related between the two. The landing r a d a r is on the descent stage, not shown, and is,
therefore, discarded on the lunar s u r f a c e a f t e r it has served i t s function during landing.
OVERALL BLOCK DIAGRAMS
The signal interconnections among the various equipments which constitute o r have
some part in the guidance, navigation, and control a r e illustrated in Fig. 11-44 and
Fig. 11-45 f o r the command module and LEM systems, respectively. The equipment and
signals shown on these figures can, f o r the most part, be related to m a t e r i a l already
discussed.

A detailed explanation is not given h e r e since the intent is only to show the

general nature of the equipment interfaces, the similarity and differences between the
command module and LEM systems, and the c e n t r a l r o l e of the guidance computer in
each case.

L.

*.

11-62

IT (IMU)

11-42

Location of the Guidance and Navigation System in the


Command Module

\AREA

FOR BACKUP
ATTITUDE REF.

11-43

Location of the Guidance and Navigation System in the


Lunar Excursion Module

11-6 3

TO SCS

ANGLE INCREMENTS IFOR IMU BACKUP I


1

FINE

ALIGN

*CAGE'

ISCT1
SCANNING

CLOCK SYNCH

ENGINE

ON / O F F

TIMING 8
TELEMETRY

TELEMETRY

GYRO TOR9

ALIGN

SIGNALS

sxT*

-.

I SPS I

TWO D I A
CDU'S

CDU'S

INERTIAL

SENSORS

DOWNLINK

MASTER

O ATTITUDE BALL DISPLAY

I
(IMU)

UPLINK

( BMAGS a GDC )
BODY MOUNTED
ATTITUDE GYROS
A N D COUPLER

SERVICE
PROPULSION
SYSTEM

'

S X T POINTING
COMMAND ERROR

FROM I M U CDU'S

SIXTEEN
REACTION J E T S

O
TW
N EL
sM
VE

It

1I

SPACECRAFT
THRUST

CDU'S
(SXT I
SEXTANT

S-IVB SEPARATE /ABORT,

FROM TWO D/A CDU'S


(CGC 1
COMMAND
MODULE
GUIDANCE
COMPUTER

OPTICS
MARK BUTTONS

MANUAL
CONTROLS

SPACECRAFT
IMPULSE
CONTROLLER

MARK

REJECT

LEM ATTACHED.

GUlD

MARK

TO IMU FOR
COARSE ALIGN*?
?YAW, *ROLL RATE COMMANDS

ROTATIONAL
HAND C O N T R O L L E R

?PITCH, ?YAW. ?ROLL RATE COMMANDS


(SINGLE LEVEL1

TRANSLATIONAL
HAND CONTROLLER

'X.

t Y , t Z . TRANSLATION

I SINGLE LEVEL I

DSKY

KEYBOARD

11-44

INPUTS

COMMANDS

CLOSURE

S..A T U R N
~

11

s-IVB
INSTRUMENT
UNIT

CUTOFF

SEQUENCER

MISCELLANEOUS
DISCRETES

RELEASE, LIFTOFF, 8 ULLAGE

INJECTION START AND 5 - I V B

'PITCH.

FROM

S M SEPARATE

SERVICE PROPULSION READY


BLOCK UPLINK
PRIMARY CONTROL. AUTO. HOLD. 8 FREE
CM CONTROL OF SATURN I BACKUP1

(BACKUPI

SATURN

I
ATTITUDE
ERROR
NEEDLES

GIMBAL ANGLES

FROM IMu

A T I Z P E

DISPLAYS

CAUTION
PGNCS

TWO C O M P U T E R
DISPLAY 8 KEYBOARD
8 RELAYS

CAUTION. CGC W A R N .

SYSTEM STATUS

NUMERICAL

CAUTION

ISS WARN
LIGHTS

8 WARNING

ON DSKY'S

READOUTS

Guidance, Navigation, & Control Interconnections in


Command Module

TO S C S
INITIAL CONDITION

DATA
TIYINb

ABORT GUIDANCE
SYSTEY

ALIGNMENT

DATA

THREE A / D
CDU's
Y E A S U R E Y E N T UNIT

I MU
.CAGE*

--

I l M U GIMBAL
ANGLES
ACCELEROMETER
SIGNALS

FINE ALIGN GYRO TORR SIGNALS


COARSE ALIGN SIGNALS
FROM

ALIGNYENT OPTICAL
TELESCOPE

Iuu.CoU*S
c

* M A R K ' SIGNALS

SENSOR

TWO

D/A
(LGCI

RENDEZVOUS
RADAR

LEY
GUIDANCE
COYPUTER

H
U

MASTER

CLOCK SYNCH

ENGINE ARMED
ENGINE O N f O F F

ENGINE
SEQUENCER

THRUST LEVEL INCREMENT COMMANDS


TRIM GIMBAL INCREMENT COMMANDS
TRIM GIMBAL FAILURE DESCRETES

RCS

JET PAIR FAILS


G B N CONTROL. AUTO. HOLD, 0 AUTO THROTTLE
ABORT 8 ABORT STAGED
DISPLAY INERTIAL DATA
BLOCK UPLINK ( N O T USED 1
HORIZONTAL VELOClTY LOW SCALE

I
1

I
I

TELEYETRY

ASCENT
EN6INE

SPACECRAFl
THRUST
B TDROUE

DESCENT
ENGINE

16
REACTION
JETS

JET O N / OFF SIGNALS

STAGE

TELEYETRY

CONFIRM

FROM PYRD

MISCELLANEOUS
DISCRETES

MANUAL
SWITCH
CLDSURE

u1
TO !MU FOR
COARSE ALIGN
LANDING
RADAR

ATTITUDE
NEEDLES

ROTATIONAL
I H A N D CONTROLLERS

PITCH YAW 8 ROLL RATE COMMANOS


( ANALOG 1
I

TRANSLATION
IHAND CONTROLLERS1

!X.

!Y,

(SINGLE

I I

ALTITUDE A N D
ALTITUDE RATE

I
DISPLAYS

LEVEL)

R A T E OF D E S C E N T
CONTROLLER

I I

SERIAL WORDS

0 2 2 TRANSLATION COMMANDS

MANUAL
CONTROLS

H 8 H DIGITAL

D f A CDU'S

VELOCITY

COMMANDS

?AH

(DSKYI
COYPUTER

'

DSKY

KEYBOARD INPUTS

11-45

RELAYS

CAUTION'
8 WARNING
SYSTEM STATUS 0 CAUTION
NUMERlCAl

LIGHTS ON DSKY

READOUT

Guidance, Navigation, & Control Interconnections in LEM

CHAPTER 11-4
OPERATION MODES O F GUIDANCE, NAVIGATION, AND CONTROL
APOLLO COMMAND MODULE BLOCK I

The s y s t e m that h a s been described so f a r can be seen to have a high degree of


flexibility in performing the many t a s k s of concern. In this chapter a s e r i e s of d i a g r a m s
a r e used to show briefly the equipment involved and the information flow in operations
with these tasks.

TO CELESTIAL BODY

,.

OPTICAL UNIT

ENTERING COMMANDS)

STABILIZATION AND
MANEUVER COUPLING UNIT

Fig. 11-47

Equipment Arrangement in the Command Module L o w e r Equipment Bay

The figure above shows the installed arrangement of the equipment in the Block I
command module.

In the following figures this equipment is shown separated in o r d e r

to t r a c e signal paths m o r e easily, The Block I equipment configuration lends itself


better to the t a s k s of this chapter than the l a t e r m o r e integrated Block I1 equipment
which w i l l finally perform the lunar landing.

11-67

OPTICS CDU'S

I DSKY)

IMU CDU'S

@
@opTlcs

CONTR
HAND

COMPUTER

r-7

%MARK

sM

BUTTON
MINIMUM
IMPULSE
SPACECRAFT
ATTITUDE HAND
CONTROLLER

Fig. 11-48

Subsystem Identification

This is the key figure of the s e r i e s .

Here the principal subsystems of the Block I

command module guidance and navigation a r e a r r a y e d and identified f o r u s e in the s u b sequent f i g u r e s , The s e n s o r s of the s y s t e m a r e shown in the top center: the two optical instruments, sextant and scanning telescope, and the inertial measurement unit
( I M U ) all mounted on the common rigid navigation base. At top left a r e the two s e t s of
coupling data units (CDUs) to provide the communication of the optics and IRlU angles

with the computer shown a t the center.


a t upper right,

The computer display and keyboard (DSKY) is

The whole vehicle - command and s e r v i c e modules

the figure center right.

is represented by

The s e p a r a t e stabilization and control s y s t e m of the Block I

system is bottom right. The astronaut navigator is shown bottom left surrounded by
s e v e r a l of the important controls.

11-68

ENGINE START

COMPUTER SOLVES "GUIDANCE


LAW" AND GENERATES
COMMANDED ATTITUDE

Fig, 11-49

RCS& SPS
CONTROL

Guidance Steering Control

This f i r s t mode is that of powered flight guidance.

The signals f r o m the accel-

e r o m e t e r s on the IMU a r e processed within the computer where the steering equations
develop a desired thrusting attitude of the vehicle to achieve the desired direction of
acceleration. This is treated as a commanded attitude which is compared in the CDUs
with actual attitude measured by the IMU. The difference is a s t e e r i n g e r r o r which is
sent to the SCS to control the vehicle, Resulting vehicle motion is sensed by the IMU to
complete the feedback, When the required velocity change is achieved, the computer
sends a rocket engine shutdown signal.

The crew can monitor the whole operation by

the display of appropriate variables on the DSKY such a s the components of velocity yet
to be gained.
Before the IMU can be used f o r an operation such as this, it must be aligned to
the desired spacial orientation.

This p r o c e s s is described next.

11-69

IMU alignment is normally performed in two stages: ' ' c o a r s e ' ' and "fine". C o a r s e

alignment is described h e r e in two steps using the figures on the opposite page.

The first s t e p of c o a r s e alignment is to give the computer a reasonably accurate


knowledge of spacecraft attitude with respect t o the celestial framework being used.
Illustrated here, the navigator sights sequentially two stars using the scanning telescope
(SCT). The s t a r image is sensed by the navigator who u s e s his left hand optics cont r o l l e r t o command the SCT p r i s m such a s to center the s t a r on the reticle. He pushes
the m a r k button when he achieves satisfactory tracking which signals the computer to
read the SCT angles being transmitted it by the optics CDUs. A second s t a r direction
a t a reasonably l a r g e angle f r o m the f i r s t is similarly measured.

The navigator identi-

fies which s t a r s a r e being used to the computer through the keyboard of the DSKY. With

these data the computer determines in t h r e e dimensions the spacecraft attitude which is
held reasonably fixed by a gyro control attitude hold of the SCS during all of these c o a r s e
alignment operations.

In s t e p two the computer determines desired IMU gimbal angles based upon its
knowledge of spacecraft attitude and the guidance maneuver which w i l l be next p e r formed. These desired angle signals sent to the IMU through the CDU a r e quickly
matched by the IMU gimbal s e r v o s in response to e r r o r signals developed on the angle
t r a n s d u c e r s on each gimbal aixs. At this point the IMU gimbal s e r v o s a r e then switched
over to the gyro stabilization e r r o r signals to hold the achieved orientation.

11-70

Fig. 11-50

IMU C o a r s e Alignment Step 1

DESIRED IMU GIMBAL ANGLES

@@

Fig. 11-51 IMU Coarse Alignment Step 2

11-71

Fine IMU alignment w i l l a l s o be described in two steps using the f i g u r e s on the


opposite page.

In the first step two s t a r directions a r e again measured by the navigator. This
time he u s e s the high magnification of the sextant (SXT) with the precision readout on
the s t a r line in o r d e r to achieve n e c e s s a r y accuracy. The IMU is presumed to be under
gyro stabilization control and t o be reasonably close to the desired orientation. On each
of two stars, which the navigator identifies to the computer, the navigator signals

" mark" when he achieves p r e c i s e alignment on the SXT c r o s s h a i r .

On these signals

the computer simultaneously reads the SXT and IMU angles being transmitted through
the CDUs.

With these data the computer determines s t a r directions in IMU stable m e m -

b e r coordinates f r o m which the spacial orientation of the IMU being held by gyro cont r o l can be computed.

The spacecraft attitude need not be held fixed during these fine

alignment operations a s long as the angular velocity is small enough to p e r m i t accurate


s t a r tracking by the navigator.

In s t e p two the computer determines the existing IMU attitude e r r o r based upon
the desired attitude a s determined f r o m the next use of the IMU such a s f o r a particular
guided maneuver. The computer then m e t e r s out the necessary number of gyro torquing
pulses necessary to p r e c e s s the gyros and the IMU to c o r r e c t the IMU alignment e r r o r .
The above two steps can b e repeated i f desired to obtain m o r e precision in the fine
alignment when the torquing precession angle is large.

11-72

,*

/',

* I

/&
/

/
:
/

"

Fig. 11-52

FREE" TUMBLE
PERMITTED

Manual IMU Fine Alignment Step 1

W T E R DETERMINES IMU
ATTITUDE ERROR AND TORQU&S
GYROS TO CORRECT ERROR

Fig. 11-53

Manual IMU Fine Alignment Step 2

11-73

Fig. 11-54

Low Orbit Navigation

Landmark Tracking

Onboard navigation measurements in low orbit can be performed either using


landmark r e f e r e n c e s as shown on the above figure o r using other r e f e r e n c e s as described
later. In the above figure, the navigator f i r s t aligns the IMU a s previously described
and then t r a c k s identified landmarks a s they p a s s beneath him using the SCT. When he

is on target he signals " mark" and the computer r e c o r d s IMU and SCT angles and time
so as to compute landmark direction in the coordinate f r a m e of the aligned IMU. These
direction m e a s u r e m e n t s a r e then used to update the computer' s estimate of position and
velocity and the computer' s estimate of e r r o r in these p a r a m e t e r s . These data can be
displayed to the astronauts if d e s i r e d ,
Although the above a s s u m e s identified landmarks of known cbordinates, unidentified landmark f e a t u r e s can be used a s described in P a r t V.

/'

CORRECTION TO

PARES MEASURED
STAR- LANDMARK

Fig. 11-55

Midcourse Navigation

Manual Star- Landmark Measurement

The u s e of the sextant to measure the angle between identified s t a r s and landmarks
f o r midcourse navigation is described with the above figure,

The acquisition p r o c e s s

using the scanning telescope is assumed already to have been performed so that the d e s i r e d star and landmark images appear i n the SXT field of view.

With h i s right hand

the navigator periodically commands jet impulses to hold the landmark in the field of
view by controlling spacecraft attitude and the body-fixed landmark line. With his left
hand he controls the sextant m i r r o r to superimpose the s t a r image onto the landmark.
When this superposition is satisfactory he signals "mark" and the computer r e c o r d s the
measured navigation angle and time. These numbers a r e then further processed i n the
computer navigation routines. The computer displays the correction to the state vector
which would be caused by this sighting so that the navigator is given a b a s i s to r e j e c t a
faulty measurement before it is incorporated into the navigation.

11-75

The use for navigation of the automatic s t a r tracker (AST) and photometer (PHO)
is shown in two steps with the figures on the opposite page.

In the first step the navigator uses the scanning telescope (SCT) to acquire the
navigation s t a r with the automatic s t a r tracker on the sextant. Acquisition is confirmed
by a "star present" light signalled from the s t a r tracker.

In step two the navigator maneuvers spacecraft attitude manually to point the bodyfixed horizon photometer line to the illuminated horizon by observing the geometry
through the SCT. The SCT has a reticle pattern which permits the navigator to judge
when the photometer is looking in the plane containing the star and the center of the
planet. This puts the photometer sensitive area directly beneath the s t a r , His task is
then to sweep the photometer line in this plane through the horizon. When the sensed
brightness drops to half the peak value, the photometer automatically sends a "mark"
to the computer so that the resulting navigation angle and time can be recorded.
This operation can be performed using the sun illuminated limb of either the moon
o r earth. Operation with the earth depends upon the systematic brightness of the atmospheric scattered light with altitude described in P a r t V.
The navigation measurement process described above uses astronaut control i n
positioning the photometer line. If the IMU is on and aligned, this process could be
completely automatic through computer control program.

11- 76

AUTOMATIC

.
R

Fig. 11-56

/'

Illuminated Horizon Manual Navigation Measurement

- Step 1

.
nu Ivmml

I-

ORRECTION TO

AND EXPECTED STAR-HORIZON


ANGLES A N D U.PDATES STATE

Fig. 11-57

Illuminated Horizon Manual Navigation Measurement

Step 2

The automatic star t r a c k e r on t h e sextant provides the capability of automatic IMU


alignment a s illustrated in two s t e p s in the figures on the opposite page. Without a s t r o naut help, however, the s t a r t r a c k e r cannot acquire a known alignment star unless the
IMU is already roughly aligned to provide a c o a r s e direction reference.

The automatic

IMU alignment capability described here, then, is most u s e f u l to r e - c o r r e c t the IMU


d r i f t after a long period of IMU operation,

In the f i r s t s t e p the computer points the sextant t o the expected s t a r direction


through the optics CDUs based upon t h e vehicle attitude measured by the IMU. P r e sumably the star t r a c k e r now s e n s e s the desired s t a r within its acquisition field of view
and signals the computer that the s t a r is detected.

The computer now changes equipment mode, step two, to send the s t a r t r a c k e r
e r r o r signals to the sextant drives so as to t r a c k the star automatically. The computer
then r e a d s simultaneously the sextant and IMU angles in o r d e r t o determine two components of the actual IMU misalignment. This l a t t e r is corrected by computer torquing
signals to the IMU gyros as described previously. Acquisition and tracking of a second
s t a r complete the automatic fine alignment in t h r e e d e g r e e s of freedom.

11-78

i ,

/*

t
0 CONTROL
IMU UNDER GYRO
D OF PRECISION
BUT IN NEED
REALIGNMENT
" STAR

PRESENT "

MIRROR COMMANDS
IMU ANGLES

ANGTE

Fig. 11-58

U S IN G I M U
DATA,
COMPUTER GENERATES MIRROR
COMMdNDS TO POINT TO AVAILABLE
STAR, WHBN STAR TRACKER DETECTS
STAR IT SIGMALS COMPUTER WHICH
INITIATES STEP 2 .

Automatic IMU Fine Alignment

- Step 1

/
AIITOMATIC
AUTOMATIC
STAR TRACKING

GYRO TORQUING

Fig. 11-59

Automatic IMU Fine Alignment

11-79

- Step 2

I
n

Fig. 11-60

"STAR PRESENT" SIGNAL


DISAPPEARS WHEN STAR
SINKS BELOW HORIZON

COMPUTER COMPARES TIME OF OCCULATION


WITH EXPECTED TIME AND UPDATES NAVIGATION
STATE VECTOR AND ERROR MATRIX

Star Occultation by Moon

Automatic Navigation Measurement

The automatic s t a r t r a c k e r provides the means for making automatic star occultation navigation measurements with the moon, a s shown above.

An acquisition by the

s t a r t r a c k e r as shown in Fig. 11-56 o r Fig. 11-58 is required, of course, a s an initial


step.

While the s t a r is being tracked, the instrument generates a " s t a r present" signal

f o r the computer which is based upon the detected s t a r light energy.

A s the s t a r sinks

below the lunar horizon due to the orbital motion of the spacecraft, t h e s t a r present
signal disappears at the moment of occultation. The time of this event is measured by
the computer as a point of the navigation data.

A s i m i l a r p r o c e s s is possible using the earth' s limb, but this r e q u i r e s a m o r e


elaborate s t a r present detection. The s t a r intensity diminishes gradually due to dispersion and scattering a s the beam sinks into the earth' s atmosphere.

11-80

CORRECTION TO
STATE VECTOR
OCCULTATION
MEASUREMENT

MPARES TIME OF
WITH EXPECTED
TIME A N D UPDATES NAVIGATION
STATE VECTOR AND ERROR MATRIX.

Fig. 11-61

Star Occultation by Moon

Manual Navigation Measurement

Besides the automatic occultation measurement just described, a manual detection


is possible, of course, as shown above,

This is of advantage since it does not r e q u i r e

that the optics system electronics be turned on. In fact, the event can be observed by
the astronaut through the window and timed with a s e p a r a t e stopwatch for transmission
to the e a r t h for use in aiding ground-based navigation measurement.

11-8 1

CHAPTER 11- 5
SPACECRAFT SAFETY CONSIDERATIONS
O F GUIDANCE, NAVIGATION, AND CONTROL

Although the risk is actually small, the Apollo crew, whenthey embark in their
spacecraft admittedly put their life in jeopardy.

However, unlike the m o r e traditional

pioneers and adventurers, the men flying the Apollo missions w i l l leave i n a spacecraft
only after their safety is assured.
preparation for the voyage.

Crew survival w i l l be a most strong concern in the

The National Aeronautics and Space Administration has s e t

high safety standards: The crew, in a checked-out vehicle leaving the earth launching
pad for the lunar surface, should have a 90% probability of accomplishing the lunar
landing mission and have a 99. 9% probability of returning to e a r t h safely whether having
been able to complete the mission o r not.

These goals a r e sought by consideration of all

p a r t s of the Apollo program: mission planning, spacecraft design, crew training, testing
methods, etc.

In this chapter we a r e concerned only with the Apollo safety aspects

under consideration in the guidance, navigation, and control systems.


Much can be said about the means of producing complex equipment which has an
extremely low failure probability.

Questions of discipline in basic design, p a r t s selec-

tion, manufacturing techniques, qualification methods, testing procedures, and other


reliability enhancing techniques a r e much debated.

We will bypass t h i s well-treated

subject and look at a s p e c t s of design and planning which accept the occurrence of failure
without the occurrence of d i s a s t e r ,
The tolerance to failure i n Apollo systems is based primarily on the deliberate d e sign guideline that any single failure should, if at all possible, leave enough working
equipment remaining to bring the crew safely home.

Although f o r practical reasons,

this guideline cannot be met everywhere, the number of safety c r i t i c a l flight i t e m s that
have no backup is quite small.
ABORT TRAJECTORIES
The guidance and navigation equipment is designed with enough flexibility in hardware and in computer program to support the measurements and maneuvers n e c e s s a r y
f o r a l l reasonable mission abort t r a j e c t o r i e s required due to failures in other p a r t s of
the spacecraft.

Depending upon the nature of this failure and the phase of the mission

failure occurs, the crew can initiate an abort by so informing the flight computer and

11-83

A I , A2 ,, A3 - Abort Points
A3

Boost Phase Abort

\
\

11-62

Near E a r t h Abort T r a j e c t o r i e s

n
0 Fuel Critical

Abort
Time Critical

3'

To
Entrv

Circumlunar
Late Abort

,<

Late
Abort
Decision

A I , A*, A3

- Abort Points

-mRepresents
Midcourse Correction

11-63

Translunar Coast Abort T r a j e c t o r i e s

11- 84

setting the proper condition of the appropriate propulsion systems. In some situations
the pilot can inform the computer which of t h r e e types of abort h e wishes: (1) t i m e
c r i t i c a l a b o r t s which r e q u i r e f a s t e s t r e t u r n using all available propulsion, ( 2 ) propulsion
critical a b o r t s which r e q u i r e optimum use of available fuel in energy efficient orbital
4

t r a n s f e r s , and ( 3 ) normal a b o r t s which use t r a j e c t o r i e s which a r e constrained to achieve


a landing on one of the prepared e a r t h recovery a r e a s .

The computer can inform the

crew about the t i m e s of flight and propulsion usage f o r each of the above a b o r t s so that
the abort mode decision can be made.
The abort trajectory to be determined and controlled by the guidance and navigation
equipment depends upon the mission phase in which the abort decision is made, Figure
11-62 i l l u s t r a t e s the t h r e e abort trajectory types pertinent to operations n e a r the earth.

Trajectory 1 on the diagram is d i r e c t abort to e a r t h during launch boost a s c e n t , It is


flown when the failure is of a nature requiring immediate r e t u r n t o earth o r where sufficient propulsion i s n ' t available t o fly trajectory 2.

Abort t r a j e c t o r y 2 continues the

flight into e a r t h orbit using a n upper stage of the vehicle.

It has the advantage of better

choice of landing recovery a r e a by selecting the phasing of the r e t u r n maneuver.

It

further p e r m i t s a possible continuation of the flight but obviously of m o r e limited m i s sion scope.

The descent f r o m orbit, t r a j e c t o r y 3, is s i m i l a r to the e a r t h orbit r e t u r n s

already flown by the Soviet and American manned orbital flights.


Aborts that can be initiated a f t e r Apollo has been committed through translunar
injection a r e illustrated i n Fig. 11- 63.

T r a j e c t o r i e s 1 and 2 on t h i s figure a r e typical of

the paths flown f o r aborts' initiated during the first part of the translunar coast.

Tra-

jectory 1 illustrates a fuel optimum d i r e c t r e t u r n to earth. T r a j e c t o r y 2 illustrates the


full fuel usage quick r e t u r n to earth. At some point in the translunar coast, the time to
earth r e t u r n is quicker i f the spacecraft coasts around back of the moon and then continues home, t r a j e c t o r y 3,

All of these cislunar aborts w i l l r e q u i r e careful navigation,

Navigation is required before the abort is initiated upon which to base the abort injection
guided maneuver.

After this maneuver, navigation is needed upon which to base s m a l l

midcourse corrections to a s s u r e accurate arrival at the safe e a r t h entry conditions.


After a r r i v a l into lunar orbit, a b o r t s either may be an immediate t r a n s e a r t h injection o r may necessarily be preceeded by recovery of the two men i n the LEM.

Figure

11-64. i l h s t r a t e s the t r a j e c t o r i e s and operations involved with the LEM aborts.

Trajec-

t o r y 1 i l l u s t r a t e s a typical abort initiated during the LEM descent,


+-

The abort trajectory

,injection, begun a t point A1, is guided and controlled to put the LEM i n a fairly high
elliptical trajectory s o that the phasing is proper for rendezvous to meet the orbiting
command module a t point R1. Midcourse corrections, not shown, a r e n e c e s s a r y based
upon navigation f r o m the rendezvous r a d a r or e a r t h tracking data, Unfortunately much
of the t r a j e c t o r y o c c u r s i n back of the moon out of sight of the e a r t h tracking facility.
This might suggest use of a low altitude holding orbit such a s described below t o provide
better phasing of the rendezvous f o r e a r t h coverage.
11-85

T r a j e c t o r y 2 of Fig. II-64illustrates a typical LEM emergency a b o r t from the


lunar surface. Here it is supposed that a failure h a s occurred - such as fuel tank leak age o r life support system failure - that r e q u i r e s immediate ascent without waiting until
the CSM is in the proper position for a n o r m a l a s c e n t and rendezvous.

The powered

phase is guided to put the LEM in a low altitude c l e a r perilune orbit where it w i l l hold
until it catches up appropriately with the orbiting command module.

At the proper point

the a s c e n t engine is fired again f o r t r a n s f e r and rendezvous using midcourse corrections


as required.

Alternately, once the LEM succeeds i n getting into a holding orbit, it can

a s s u m e , i f n e c e s s a r y , a passive r o l e and allow the CSM to maneuver f o r rendezvous


and crew pickup.
CONTROL OF PROPULSION FAILURE BACKUPS
The Apollo guidance and control equipment w i l l be designed to operate with abnormal propulsion and loading configurations f o r given mission phases to provide abort
capabilities covering f a i l u r e in any of the p r i m a r y rocket engines. F i g u r e 11-65 is a
r a t h e r fanciful diagram showing examples of a b o r t s of this nature. The heavy ascending
line t r a c e s out the normal mission phases f r o m prelaunch to lunar orbit. The dashed
lines t r a c e out abort paths through alternate propulsion s o u r c e s to cover f a i l u r e s of the
normal rocket used i n each phase. These paths a r e numbered on the diagram and a r e

explained briefly below:


1.

This is the u s e of the launch escape s y s t e m providing a b o r t s during the period


f r o m on the pad before liftoff until atmospheric exit during the e a r l y p a r t of
the second stage burn. No measurement is n e c e s s a r y by the guidance s y s t e m
f o r launch escape a b o r t s ; the s y s t e m is designed to pull the command module
safely past and far enough away f r o m a n exploding booster f o r a low velocity
entry and normal CM parachute landing.

2.

A failure of the second stage during ascent might be of a nature to allow


thrusting i n the third SIVB Saturn stage into e a r t h orbit. This would naturally
deplete SIVB fuel sufficiently to prevent continuation of a lunar mission.

3.

Again during second stage boost and during third stage as well, the abort may
be made to an immediate entry t r a j e c t o r y and landing using the command
module propulsion and the spacecraft guidance and control s y s t e m s .

4.

Aborts using service module propulsion during third stage boost may a l s o be
made into e a r t h orbit. A second burn of the s e r v i c e module would then initiate
descent to a selected landing site.

5.

If the abort is initiated while in e a r t h orbit the s e r v i c e module propulsion


would be used f o r descent assuming it still functions. If not, the s m a l l reaction
j e t s could be used i n a limited r e t r o g r a d e translational burn o r s e r i e s of b u r n s

so as to capture the atmosphere.

11-86

@Abort during landing


L

@Emergency abort
from lunar surface
Heavy line is powered
maneuver +"-

A1,

'' R

- Abort points
- Rendezvous

I
I

points

\
\

To
Earth

11-64

LEM Operations Abort T r a j e c t o r i e s

,.--.,

Earth Orbit

S-IVB
d a n s l u n a r

-..

/'

/'

,
K e y to Propulsion Used:
1st stage

s- I C

Boaster

s-IC

S- IVB 3rd stage


LES Launch Escape System
SM Service Module

L E M Lunar Excursion Module

Prelaunch

11-65

Recovery

@, @, etc See text for description I

Propulsion F a i l u r e Abort Paths

11- 87

"OR" sum

1
gyro error
Inertial
Measuring
Unit
and
Associated
Systems

power supply fail

-+@

Computer

-a
@
-a

accelerometer error
C D U error

A A

@ master inertial error

~
~

-b

-+

Alarm
Program
&

Error Detect
Circuits

w etc.
~

manual
reset input
11-66

computer e,rror
inertial attitude fail
accelerometer fail

4-w f

CDU fail

master guidance fail

detect

condition
display lights

Failure Detection - Command Module P r i m a r y Guidance System

6.

On the way to the moon the s e r v i c e module propulsion could be used t o inject
into the r e t u r n o r b i t s described previously.

7.

If the s e r v i c e module rocket h a s failed the flight can continue around the moon
on the "free return'' path using the reaction j e t s in translation maneuvers to
perform the n e c e s s a r y midcourse maneuvers determined by navigation.

8.

If s e r v i c e module propulsion fails while in lunar orbit before the LFM s e p a r a tion and descent, the LEM propulsion and LFM guidance and control s y s t e m s
can be used to inject the command module onto the n e c e s s a r y t r a n s e a r t h
trajectory.

These examples of propulsion failure abort paths illustrate dramatically the


n e c e s s a r y flexibility and universality needed of the Apollo guidance, navigation, and
control systems.
FAILURE DETECTION AND ALARM

A central aspect of mission safety is the early detection of s y s t e m failure. Part


of this detection is in the systematic onboard testing during t h e s t r e s s - f r e e coasting
phases t o a s s u r e the needed s y s t e m s a r e functioning. Of m o r e interest, perhaps, is
the automatic failure detection f e a t u r e s which iunmediately signal appropriate a l a r m s
during the s t r e s s e d accelerated phases of rocket thrust o r entry. It is with the help of
these a l a r m s that the crew can initiate appropriate abort action immediately as necess a r y . A s a n example, we w i l l limit this discussion to the automatic failure detection of
the guidance system i n the command module a s used during guided flight.
F i g u r e 11-66 is a simplified diagram of this failure detection system.
the left r e p r e s e n t s the inertial system.

The box at

The signals coming out a r e e r r o r detections.

Shown is ''gyro e r r o r " which is a signal which exists when any of the gyro gimbal stabilization loop s e r v o e r r o r s exceed a preselected detection level.

The " accelerometer

e r r o r " and "CDU e r r o r " have s i m i l a r properties with detection of any of the a c c e l e r ometer s e r v o loop e r r o r s and coupling display unit s e r v o loop e r r o r s ,
The "power
supply fail" signals deviations of the i n e r t i a l s y s t e m power supply voltages f r o m p r e selected levels.

P a c h of these detections is sent to the computer as well a s being

separately summed t o light a m a s t e r inertial s y s t e m e r r o r display light. During system


turn-on o r mode switching this light is expected to operate briefly during the transient,
but w i l l extinguish itself i n a normal system,
The computer contains its own e r r o r detection p r o g r a m s and circuitry which may
light a m a s t e r computer e r r o r light.

If t h i s o c c u r s f o r transient reasons, the astronaut


w i l l succeed in extinguishing it by pushing a r e s e t button. The computer program w i l l
examine its e r r o r and the inertial system detections and w i l l light appropriate fail lights.
The " inertial attitude fail" signals that the inertial system alignment is lost and the crew
should use a backup system if re-alignment cannot be accomplished. The " accelerometer

11-89

'

PRIMARY
I SYSTEM

L
_"""" (

VEHICLE
VELOCITY
1-1
INCREMENTS

I
"'^'\PUTER
I
~1
Lu'v

I
I

4b

D O W N : BACKUP

VEHICLE

"_"

"
"
"

- - ---

"
"
"
"
"

"

"
"

JDYNAMICS

" I 1

"
"
"

CT,

UP: PRIMARY

Ii

I
H
H
I

r"

v
THRUST
*AX'S
BODY M,OUNTED
ACCELEROMETER

THRUST AXIS
V E L OCITY

I
I
I
I

b COUNTER

INCREMENTS
A

GIMBAL

11- 67

Primary & Backup Guidance Concept

I'

fail" indicates that the acceleration data in the guidance is faulty and the p r i m a r y guidance steering cannot be used.

In t h i s latter case, however, the inertial attitude data

may still be c o r r e c t for use in a backup mode.

A s i m i l a r situation o c c u r s with the

"CDU fail" light.


The l a s t light is a " m a s t e r guidance fail" which h a s special f e a t u r e s which makes

it fail- safe.

The computer program examines periodically at a fixed frequency all of

the previous failure detections and if it finds none, the program sends out a pulse of a
particular duration.

If this pulse keeps occurring at the expected frequency, then the

detector inhibits lighting the " m a s t e r guidance fail".

Otherwise t h i s signal lights up.

If any of these lights operate the crew is trained t o take appropriate emergency
backup action.
NAVIGATION RPDUNDANCY
The originally stated p r e m i s e that a single failure should leave the system with
enough capability to r e t u r n safely should now be examined.

With r e s p e c t t o navigation

this is quite straightforward with the use of both onboard and ground-based provisions.
If ground tracking navigation data a r e unavailable because of a loss of communications, then the onboard s y s t e m can perform all the n e c e s s a r y navigation,
If thz onboard navigation capability fails, the ground can provide the necessary
data. This applies to the failure of either the optics s y s t e m o r the onboard computer.
F o r a failed computer, onboard navigation data can be telemetered for ground processing

If the optics s y s t e m d r i v e s have failed, star occultation events can be observed, either by spacecraft
attitude aiming of the optics o r directly through the window. These events a r e useful
navigation data f o r the computer.
to aid in the ground determination of the n e c e s s a r y a b o r t maneuver,

GUIDANCE ANDCONTROLBACKUP
Failure in the p r i m a r y guidance equipment r e q u i r e s the use of a n alternate backup
system to provide safe r e t u r n of the crew.
in Fig. 11-67.

The redundancy concept used is illustrated

The figure h a s been simplified t o illustrate m o r e clearly the principles

involved.
The p r i m a r y s y s t e m involving the inertial measurement unit gimbal system, the
flight digital computer, and the associated coupling data unit, constitutes a complete,
flexible, accurate, and fuel efficient guidance and control having the capability to p e r x

f o r m a l l the maneuvers required t o complete the mission.

The backup system is simpler,

s m a l l e r , and of m o r e modest endowment, being able to make the m o r e simple maneuvers


to r e t u r n the crew after the failure of the p r i m a r y system has aborted the mission.
Figure 11-67 shows a backup using t h r e e body-mounted single-degree-of-freedom

11-91

integrating gyros which provide attitude e r r o r signals over a limited angular range.
These e r r o r s a r e treated as steering commands t o a simple autopilot to hold vehicle
direction fixed during an abort thrusting maneuver.

Engine cutoff is signalled by the

integrated output of a single body a x i s a c c e l e r o m e t e r mounted with its sensitive a x i s


along the nominal thrust axis.

The integrator consists of a simple p r e s e t counter giving

as an output a n engine shutdown signal when the sum of the a c c e l e r o m e t e r output velocity
increments r e a c h e s a level equal to the total velocity change d e s i r e d of the maneuver.
The crew, using data telemetered f r o m the ground, w i l l pre-align the t h r u s t a x i s in the
required abort maneuver direction by aiming the vehicle with r e s p e c t to the stars.
The ground instructions to do this w i l l recognize the expected offset of the t h r u s t a x i s
f r o m the vehicle r o l l axis. The maneuvers a r e r e s t r i c t e d to accelerations i n a fixed
direction but of any magnitude s e t into the acceleration counter. If l a r g e magnitude
velocity changes a r e required, then the m o r e limited a c c u r a c y of this backup s y s t e m
w i l l r e s u l t i n significant e r r o r s .

These can be corrected by a much s m a l l e r maneuver

a f t e r a s h o r t coast based upon the ground tracking of the a b o r t trajectory.


This backup guidance system just explained is a somewhat simplified description
of that used i n the command module, The L E " h a s been given a m o r e complex abort
guidance s y s t e m to p e r f o r m m o r e accurately the c r i t i c a l and complex abort maneuvers
n e a r the m o o n ' s surface. This LEM abort system consists of t h r e e body-mounted r a t e measuring gyros, t h r e e body-mounted a c c e l e r o m e t e r s , and a small computer to p e r f o r m
n e c e s s a r y transformations of gyro and a c c e l e r o m e t e r data.

This computer a l s o g e n e r -

a t e s steering commands appropriate f o r abort at any phase of LEM operation to rendezvous conditions with the orbiting command module.
Satisfactory vehicle control a l s o r e q u i r e s t h r e e - a x i s spacecraft torquing during
the f r e e - f a l l and accelerating phases.
engine gimbals is provided.

Optimum redundancy in the hardware to drive the

The sixteen reaction j e t s on the s e r v i c e module and the

sixteen reaction j e t s on the LEM allow a limited number of j e t f a i l u r e s without unacceptable loss of rotational control o r translational control. Likewise the n e c e s s a r y
rotational control of the command module i s provided by a redundant assembly of 12
reaction j e t s f o r u s e during e a r t h atmospheric entry.

Various levels of automatic,

semiautomatic, and manual control can be selected by the crew to utilize the subsystems
available and working.
In the limit, the pilot o r a surviving companion can use d i r e c t hand control commands to the reaction j e t s and the engine gimbals and a view of the stars as reference
directions to provide him with the last level of emergency backup.

11-92

PART I11
EXPLICIT AND UNIFIED METHODS OF
SPACECRAFT GUIDANCE

D r . Richard H. Battin

DR. RICHARD H. BATTIN


Deputy Associate Director, Instrumentation Laboratory
Massachusetts Institute of Technology

D r . Richard H. Battin, Deputy Associate Director of Instrumentation Laboratory,

Massachusetts Institute of Technology, is a specialist i n space t r a j e c t o r i e s , guidance


concepts and in the programming of space guidance computers.
He is in charge of the Space Guidance Analysis Group which is charged with both
theoretical analysis and programming of the on-board guidance computer for the Laborat o r y ' s development of the guidance s y s t e m that will be used aboard the Project Apollo
spacecraft.

D r . Battin was b o r n in Atlantic City, N. J . , March 3, 1925, and was graduated


from F o r e s t Park High School, Baltimore, Md., in 1942. He received the S. B. degree
in electrical engineering f r o m M. I. T. in 1945 and s e r v e d in the Navy as a Supply Corps

officer f o r one y e a r . He returned to M. I. T. i n 1946 as a n instructor in mathematics


and a r e s e a r c h a s s i s t a n t in meteorology, studying atmospheric circulations, He was
awarded the Ph, D. d e g r e e in applied mathematics f r o m M. I. T. in 1951.
Dr. Battin joined the Instrumentation Laboratory i n 1951 as a r e s e a r c h mathematician working on fire control and inertial navigation s y s t e m s and l a t e r became head
of the L a b o r a t o r y ' s Computing Devices Group.

F r o m 1956 t o 1958, D r . Battin was a senior staff member in the Operations R e s e a r c h Group of A r t h u r D. Little, Inc. , working on digital data processing f o r business
and industrial r e s e a r c h . Since returning to the Instrumentation Laboratory in 1958, D r .
Battin h a s been engaged p r i m a r i l y in interplanetary navigation theory and development.
D r . Battin is the author of the book, Astronautical Guidance (New York; McGrawHill, Inc., 1964) and co-author of the book, Random P r o c e s s e s in Automatic Control
(New York; McGraw-Hill, 1956). He h a s published numerous a r t i c l e s in professional
journals dealing with meteorology, analog and digital computing techniques, s,tochastic
processes, interplanetary t r a j e c t o r i e s and navigation. He is a member of Sigma Xi
and a n Associate Fellow of the American Institute of Aeronautics and Astronautics.

P a r t 111
EXPLICIT AND UNIFIED METHODS O F SPACECR.AFT GUIDANCE

Of fundamental importance in the design of space guidance s y s t e m s is the creation


of both flexible techniques and versatile instrumentation which have a wide range of
applicability but neither compromise mission accuracy nor place an undue burden on
propulsion requirements. Minimal constraints on the system and methods of i t s operation should be imposed by detailed mission objectives which a r e subject to frequent
and l a s t minute revision. In partial fulfillment of these goals, the development of explicit guidance techniques is warranted t o reduce any dependence on precomputed r e f e r e n c e orbits or specific rocket engine characteristics.
During the evolution of a space flight program such as Apollo, the ultimate m i s sion objective i s attained progressively by a s e r i e s of intermediate flights.

Each suc-

cessive flight i s planned as a direct extension of the previous one so that the need for
special equipment and untried techniques can be minimized. The s u c c e s s of this approach is enhanced through the development of unified guidance methods. Then the
guidance r e q u i r e m e n t s for each new mission phase can be met as a specific application
of a general guidance principle.
The two fundamental t a s k s of a guidance system a r e to maintain accurate knowledge of spacecraft position and velocity and to provide s t e e r i n g commands for required
changes in course.

It is the purpose h e r e to review some of the c u r r e n t techniques for

solving the guidance problem emphasizing those methods which a r e consistent with the
explicit and unified philosophy of design.

111-3

Chapter 111-1
ACCELERATED FLIGHT NAVIGATION

The t a s k of periodic determination of position and velocity, customarily r e f e r r e d


t o as navigation, divides naturally into two p a r t s

- accelerated flight and coasting flight.

F o r navigation during an accelerated maneuver, the system frequently includes inertial


instruments capable of measuring thrust acceleration along t h r e e mutually orthogonal

A guidance computer is then required to perform accurate


integrations and gravity calculations on a r e a l - t i m e basis.
a x e s which a r e nonrotating.

A functional diagram of the b a s i c computations required of the navigation s y s t e m


is shown in Fig. 111-1. Incremental outputs f r o m inertially stabilized integrating a c c e l e r o m e t e r s , together with components of gravitational acceleration computed a s functions
of inertial position in a feedback loop, a r e summed t o give the components of inertial
velocity.

The ultimate precision attainable i s , of course, limited by the accuracy of the

inertial instruments, the speed of the guidance computer and the knowledge of the initial
conditions.
GRAVITY COMPUTER
The gravity calculations may be performed in a straight-forward manner.

The

equations of motion for a vehicle moving in a gravitational field a r e

(111-1)

where r and v a r e the position and velocity v e c t o r s with r e s p e c t t o a n inertial f r a m e of


reference.

The measured acceleration vector

aT of the vehicle is defined to be the ve-

hicle acceleration resulting from the sum of rocket thrust and aerodynamic forces, if
any, and would be z e r o if the vehicle moved under the action of gravity alone. The vect o r sum of
and g, the gravitational vector, r e p r e s e n t s the total vehicle acceleration.'

aT

A simple computational algorithm, by means of which position and velocity a r e


obtained as a f i r s t o r d e r difference equation calculation, follows

-r (t,)

The vector

r (tn-l)
-

+ v ( t n m l ) At + 1 -n-1
g
(At)2 f

1Av-a( tn)

At

(111-2)

xa is the time integral of the non-gravitational acceleration forces,

the com-

ponents of which a r e the outputs of the t h r e e mutually orthogonal integrating a c c e l e r o m e t e r s as shown in Fig. 111-2.

The gravitational vector

gn is a function

of position at

t i m e tn. In the figure, only a simple spherical gravitation field is considered.


Since velocity is updated by means of the average effective gravity over the interv a l of one t i m e step, this method h a s been termed the "average g" method.

A careful

e r r o r analysis of a vehicle in e a r t h orbit has shown this algorithm t o yield e r r o r s of


the o r d e r of 100 feet and 0. 2 feet per second after a period of 35 minutes using a 2 s e c ond time s t e p and rounding all additions to 8 decimal digits.

The e r r o r s will i n c r e a s e

f o r a s m a l l e r time s t e p due to the effects of accumulated round off e r r o r s and will a l s o


i n c r e a s e for l a r g e r time steps a s truncation e r r o r s become significant.

When com-

pared t o typical a c c e l e r o m e t e r s c a l e factor e r r o r s , the e r r o r in the computational algor i t h m s e e m s t o be s e v e r a l o r d e r s of magnitude s m a l l e r ,


BODY-MOUNTED SENSORS
During r e c e n t y e a r s increasing attention h a s been devoted t o the so- called "gimballess inertial measurement unit" in which the inertial s e n s o r s a r e mounted directly to
the spacecraft (see Bumstead and Vander Velde") and Wiener(2)). Although many advantages might a c c r u e in t e r m s of s y s t e m weight, volume, power, cost, packaging
flexibility, reliability and maintainability, the realization of a satisfactory design is not
without significant problems.

Unlike the environment provided by a gimballed system,

the body mounted inertial instruments a r e subjected to substantial angular velocity which
tends t o exaggerate performance e r r o r s . Also, the r o l e of the guidance computer is
expanded since the angular orientation of the vehicle must a l s o be determined by integration of measured angular velocities. It is most convenient if the outputs of the body
mounted a c c e l e r o m e t e r s a r e immediately transformed into an inertially stabilized coordinate f r a m e so that the navigation o r guidance problem can be solved just as if a
physically stabilized platform had been employed.

A s indicated in Fig. In-3, the body fixed coordinates and the inertial coordinates
of the thrust acceleration vector a r e related by a transformation m a t r i x of direction cosines, The additional computations required of the guidance computer involve the updating of the m a t r i x and using it to t r a n s f o r m v e c t o r s f r o m one f r a m e of r e f e r e n c e t o the
other,

The transformation m a t r i x R is readily shown to satisfy a f i r s t o r d e r differential

111-6

d Z

ao

"

A
r"""""""""""""""""""""

I
I

'

I
I

4-

I
I
I
I
I

I
I

I
I

I
I
I
L"_

I
I
I
I
I
I
I

. .

a
n
Z
a

111-7

i i
+E

Cnl

-1-

rri

II

II

H
H
H

II

II

111-8

Lu

31

0
2
3"

3-

am
It

II

c:

3-

I11-9

equation with a coefficient matrix 0 whose elements a r e the components of the angular
velocity of the body fixed coordinate f r a m e measured in body coordinates.
Currently, pulse-torqued integrating gyros a r e the most promising candidates for
angular velocity s e n s o r s . However, since their basic output consists of angular i n c r e ments r a t h e r than angular velocity, the accuracy with which the transformation matrix
differential equation may be integrated is adversely affected.

The use of a higher o r d e r

integration rule provides no advantage over simple rectangular integration since the
basic data f r o m the gyro has already an uncertainty of the o r d e r of the square of the
gyro quantization e r r o r .
The accuracy attainable by a gimballess inertial s y s t e m is limited p r i m a r i l y by
the maximum angular velocities to which the vehicle is subjected.

The required s a m -

pling t i m e of the integrating gyros is inversely proportional t o this maximum angular


velocity and the time step used for integrating the direction cosine differential equations
must be of the same o r d e r of magnitude as the gyro sampling t i m e , If the sample time

is v e r y short, a digital differential analyzer may prove to be the best solution to the
problem of selecting a guidance computer.

On the other hand, if the sampling t i m e is

long enough to permit the use of a general purpose computer, t h e r e may be sufficient
t i m e remaining i n which to p r o c e s s the navigation a n d / o r s t e e r i n g equations. This is,
of course, m o r e satisfactory, for then one has the possibility of satisfying all o r most
of the complete system computation requirements with a single computer.

111-10

Chapter 111-2
COASTING FLIGHT NAVIGATION

Spacecraft navigation during prolonged coasting flight is performed by appropriate


utilization of periodic measurements of convenient physical quantities such a s (1) d i s tance, velocity, elevation and azimuth f r o m well- established r e f e r e n c e points,(Z) angles
between lines of sightto known celestial objects, (3) s t a r occultations, and (4) apparent
planet diameters.

Since navigation measurements a r e m o r e accurately made when the

s e n s o r s a r e in proximity to the data source, vehicle-borne and ground-based instrumentation can s e r v e in complementary roles.
COMPAR.ISON O F METHODS
The data processing a s p e c t s of the navigation problem have been the subject of
much r e s e a r c h during r e c e n t y e a r s . The classical method of the astronomer, called
the "method of differential corrections" , is cumbersome for l a r g e amounts of o b s e r vational data and is not w e l l suited t o implementation i n a vehicle-borne computer.
Blackman, in a r e c e n t paper(3), gives an excellent review of s e v e r a l of the new methods
contrasting them with the classical approach and with each other.
Currently, the statistical methods of optimum linear estimation theory s e e m t o
hold the most promise.

The statistical method of maximum likelihood, which is based

on the concept of maximizing a particular conditional probability, has received much


attention. Optimum filter theory, whose goal is to find a linear estimator that minimizes
some function of the variances and covariances of the uncertainties in the estimated state
vector, provides an alternate method of attack. Although the subject may be approached
f r o m a variety of points of view, Potter and Stern(4) have shown that all such methods
lead to equivalent r e s u l t s if the measurement uncertainties have Gaussian distributions.
RECURSIVE NAVIGATION
The scope of this chapter does not permit a thorough development of the mathem a t i c s underlying the so-called " r e c u r s i v e method" of spacecraft position and velocity
estimation; however this is adequately treated in r e f e r e n c e (5). The method, which is
particularly well-suited to both vehicle-borne and ground-based computation, is under
active consideration for use in the Apollo guidance computer as well as in the Mission
Control Center a t the Manned Spacecraft Center, Houston, Texas.

I11-11

The estimate of position and velocity is maintained in the computer in non-rotating


rectangular coordinates and is referenced to either the e a r t h o r the moon.

An e a r t h

centered equatorial coordinate s y s t e m is used when the vehicle is outside of the lunar
sphere of influence.

Inside of this sphere the c e n t e r of coordinates coincides with the

center of the moon.

The extrapolation of position and velocity is made by a direct nu-

m e r i c a l integration of the equations of motion.


The basic equation may be written in vector f o r m as

(111-3)

rPV

is the vector position of the vehicle with r e s p e c t to the p r i m a r y body P which


is either the e a r t h or moon and p P is the gravitational constant of P. The vector
is
the vector acceleration which prevents the motion of the vehicle f r o m being precisely a
conic with P at the focus.

where

ad

If ad is s m a l l compared with the c e n t r a l f o r c e field, direct use of Eq. (111-3) is

A s an alternative, the integration may be accomplished employing the technique of differential accelerations suggested by Encke.
inefficient.

At t i m e to the position and velocity vectors -rP V (t0 ) and x p v ( t o )


define an osculating orbit. The vector difference 6 (t) satisfies the following differential
Encke's Method

equation
3

d2
PP
6 = --dt2

PV(C)

PV(C) [ ( 1 -

) .pv-i]+ad

(111-4)

subject to the initial conditions

-s (to) =

where PV(C) is the osculating conic position vector.


would a r i s e f r o m the evaluation of the coefficient of

The numerical difficulties which

rPVin Eq.

(111-4) may be avoided.

Since
(111-5)

111-12

it follows that
"1

3
PV(C)
3

= - f(q

PV

1 - (1 + qc) 312

where
(111-6)

3 + 3 q + q2
+(l+q)3/2

f(q) = q

(111-7 )

Encke's method may now be summarized as follows:

A.

Position in the osculating orbit is calculated f r o m

(111-8)

where

C 0Y =

vpv (to)2

"

PV

(111-9)

I-lP

and x is determined as the root of Kepler's equation in the f o r m

(111-10)

The special transcendental functions S and C a r e defined by


S(x) =

1
I

3!

X
"

5!

-- x +

C(x)= 1
2!

4!

111-13

X
+ -

7!
2

x
6!

...
(111-11)

- ...

B.

Deviations f r o m the osculating orbit a r e obtained by a numerical integration


of
(111-12)
The f i r s t t e r m on the right hand side of the l a s t equation must remain
small, i. e. of the s a m e o r d e r as %(t), i f the method is to be efficient. A s
the deviation vector 6 grows in magnitude, this t e r m w i l l eventually i n c r e a s e
in s i z e .

Therefore, in o r d e r to maintain the efficiency, a new osculating

orbit should be defined by the t r u e position and velocity.

The process of s e -

lecting a new conic orbit from which to calculate deviations is called rectification. When rectification occurs, the initial conditions of the d i f f e r e n t i d

C.

equation for 6 a r e again z e r o and the right hand side is simply the p e r t u r bation acceleration Q at the time of rectification.
The position vector EPv(t) is computed f r o m Eq. (111-5) using Eq. (111-8).
The velocity vector
is then computed as

xpv(t)

(111- 13)
where

(111-14)

Disturbing Acceleration- The form of the disturbing acceleration


depends on the phase of the mission.

a r i s i n g from the non-spherical shape of the e a r t h need be considered.


lunar and t r a n s e a r t h flight,

used

During t r a n s -

the gravitational attraction of the sun and the secondary

body Q (either e a r t h or moon) a r e relevant forces.

In lunar orbit, it may be n e c e s s a r y

to consider forces a r i s i n g f r o m the non-spherical shape of the moon.


various c a s e s appears below.
A.

% to be

In e a r t h orbit only the gravitational anomalies

E a r t h Orbit

111-14

A s u m m a r y of the

where

Pi (cos 4) = 3-1 ( 7 cos $5 Pi - 4 Pi)


P; (cos

1
4) = 4-

(9 c o s $5 P'
4

Pi)

a r e the derivatives of the Legendre polynomials;

4 = L Ev -iz

cos

is the cosine of the angle

between the unit v e c t o r i E V in the direction of

rEv

of
and the unit v e c t o r i z in the direction of the north pole; r
is the
eq
equatorial radius of the earth; and J2, J3, J4 a r e the coefficients of the s e c ond, third and fourth harmonics of the e a r t h ' s potential function. The subs c r i p t E denotes the center of the e a r t h as the origin of coordinates.
B.

Translunar and T r a n s e a r t h Flight

where the s u b s c r i p t s Q and S denote the secondary body and the sun, r e s pectively.

Thus, for example,

r e s p e c t t o the p r i m a r y body.

rPSis the position vector of the sun with

The arguments q

0 a r e calculated f r o m
(111-17)

and the function f f r o m Eq. (111-7).


E p h e m e r i s data for the positions of the moon relative to the e a r t h

rss

rEM

and the sun relative t o the earth-moon barycenter


a r e required a s functions of time. The position of the sun relative to the p r i m a r y planet
is

rPs

then computed f r o m

(111-1 8 )

I11-15

In the vicinity of the lunar sphere of influence a change in origin of coordinates is made.

C.

Thus

Lunar Orbit

where A, B, C a r e the principal moments of inertia of the moon, r m is the


r a d i u s of the moon, C' is C divided by the product of the m a s s of the moon

&, -c

i a r e the selenographic coordinate unit


and the s q u a r e of i t s r a d i u s , i E ,
vectors, andJMV is the unit vector in the direction of

Navigation Measurements

- Periodically,

rMv.

the position and velocity of the space-

craft must be brought into accord with optical o r r a d a r observations made with either
onboard o r ground based s e n s o r s .

At the t i m e a measurement is made, the best esti-

mate of spacecraft position and velocity is the extrapolated estimate maintained in the
computer and denoted by 2 pv and 1 pv as shown in Fig. 111-4. From this estimate,
it is possible to determine an estimate of the quantity to be measured such a s an angle,
range f r o m a tracking station or range rate. When the predicted value of this m e a s u r e ment is compared with the actual measured quantity, the difference is used to improve
the estimated position and velocity vector.

A.

The Measurement Geometry Vector


An important feature of the r e c u r s i v e navigation method is that m e a s u r e ment data f r o m a wide variety of s o u r c e s may be incorporated within the same
framework of computation. Associated with each measurement is a six-di-

mensional vector b representing, to a f i r s t o r d e r of approximation, the v a r i a tion in the measured quantity q which would result f r o m variations in the cornponents of position and velocity. Thus, each measurement establishes a com-

ponent of the s p a c e c r a f t state vector along the direction of the b vector in state
space,

I11 -16

'7M

>

h > In

&I

111-17

d
C

*I
$"

cd

=r
a

E0
u

Specifically, if b and b are the upper and lower three-dimensional


-1
-2
partitions of the six-dimensional b vector and if 6 q is the difference beN

tween the value of the quantity as actually measured and the expected value
as computed f r o m the c u r r e n t values of

where 6

r pv and 1 pv,

then

r pv and 6 pv a r e the changes in the computed values of

position

and velocity n e c e s s a r y t o make the estimated state of the vehicle compatible


with the observation.

A s examples of both ground-based and onboard measurements we may


list the following.
1. Radar R.ange Measurement

b =iRV
-1

where rRVis the range of the vehicle f r o m the r a d a r site a n d L R V is a


unit vector in the direction of the vehicle f r o m the site.
2.

Radar Range Rate Measurement

b = i
- 2 -RV

= IR.V'
where
3.

iRV

vR,vis the velocity of the vehicle with r e s p e c t to the r a d a r

site.

Star- Landmark Measurement

where r
is the distance of the vehicle f r o m the landmark a n d i L V is
LV
a unit vector in the direction of the vehicle f r o m the landmark. The unit

111-18

vector i

-S

gives the direction of the s t a r .

The notation UNIT( ) indicates

a unit vector in the direction of the quantity (


4.

1.

Star-Horizon Measurement

b2 =
-

rPV

is a vector f r o m the selected planet to the vehicle and rpH


is the altitude of the horizon f r o m the center of the planet. If the far
horizon is chosen f o r the measurement, then rPH must be negative.

where

B.

The Error Transition Matrix


Six measurements made simultaneously would provide a s e t of six equa-

tions of the f o r m of Eq. (111-21). If the directions of the associated b vectors span the state space, then the vector changes 6 pv and 6 1 pv could
be obtained by inversion of the six-dimensional coefficient m a t r i x each of
b vectors.
whose rows were elements of the Both the problems of simultaneous m e a s u r e m e n t s and m a t r i x inversion
can be avoided in such a manner that measurement data may be incorporated
sequentially as it is obtained.

For this purpose, it is n e c e s s a r y to maintain statistical data in the guidance computer in the form of a six-dimensional correlation m a t r i x E ( t ) of estimation e r r o r s . If E (t) and u (t) a r e the
e r r o r s in the e s t i m a t e s of the position and velocity vector, respectively,
96
then the six-dimensional correlation matrix E(t) is defined by

(III-22)

The t r a n s p o s e of a vector or a m a t r i x is denoted by a s u p e r s c r i p t T.

111-19

Because of accurnulated numerical inaccuracies, it is possible that this


correlation matrix may fail to r e m a i n positive definite after a l a r g e number
of computations as it theoretically must.

A recent innovation to avoid this

problem, which has a l s o the advantage of significantly reducing the total


computational requirements, is t o replace the correlation matrix by a m a t r i x W(t), called the e r r o r transition matrix.

The W(t) m a t r i x has the pro-

perty
E(t) = W(t) W(t)T

(111-23)

and thus, in a sense, is the square root of the correlation matrix. If needed,
the correlation matrix may be determined a s the product of the matrix W(t)
and i t s transpose, thereby guaranteeing it to be a t least positive semi-definite,
Extrapolation of the m a t r i x W(t) is made by direct numerical integration
of the differential equation

I O

I \
(111-24)

where G(t) is the three-dimensional gravity gradient matrix.

The m a t r i c e s

I and 0 a r e the three-dimensional identity and z e r o m a t r i c e s , respectively.


If the W m a t r i x is partitioned a s

W =

w-l

-o2
do
-2
dt

....

....

(111-25)

dt

then the extrapolation may be accomplished by successively integrating the


vector differential equations

(111-26

The G(t) matrix for translunar and t r a n s e a r t h flight is readily shown


to be

111-20

so that the differential equations f o r the w


-i ( t ) v e c t o r s a r e simply

i =1,2,
C.

...,6

The Weighting Vector

By algebraically combining t h e W matrix, the b vector and a meansquared a p r i o r i estimation e r r o r a2 in the measurement, t h e r e a r e pro-

duced a weighting vector w and the s t e p change to be made in the e r r o r t r a n sition m a t r i x t o reflect the changes in the uncertainties in the estimated
w has six
quantities a s a r e s u l t of the measurement. The weighting vector components and is determined so that the observational data is utilized in

a statistically optimum manner.


diagram in Fig. 111-5,

The required calculation is given as a f l o w '

The computation may be conveniently organized in t e r m s of the vector


partitions of the W m a t r i x as given in Eq, (111-25). If w and w2 a r e the
-1
three- dimensional upper and lower partitions of the weighting vector, then
we have

dw

zi

-i
= w i bl + -

dt

b
-2

i =1,2,

...,

(111-28)

'i 0-i

i=1

w2=
-

i=1

'w_
i
dt

Finally, the navigation p a r a m e t e r s a r e updated according to

I11-21

TRANSITION MATRIX

".

GRAVITY
GRADIENT
MATR I X

-r P"

+I 'I

MEASUREMENT
GEOMETRY VECTOR

WEIGHTING VECTOR

z =bTW
w = wr

ESTIMATED MEANSQUARED ERROR IN


MEASUREMENT

a*

z2+

W W T =CORRELATION

MATRIX OF
ESTIMATION ERRORS

STEP CHANGE IN ERROR


TRANSITION MATRIX

Fig. 111-5

Weighting Vector Calculation

POS-POS
VEL POS

POS-VEL
VEL - VEL

(111-29)
o

-i
do
-i
dt

-yz.w + a
1-1
-i
do
-i
Y ziz2
dt

"

i =1,2,.,.6

where

1 +
D.

J p 2

Numerical Integration
The extrapolation of navigation p a r a m e t e r s r e q u i r e s the solution of
seven second o r d e r vector differential equations, specifically Eqs. (111-12)
and (111-27). These a r e all special c a s e s of the f o r m
d2

7 2 = f(;r>
dt

(111-3 0 )

in which the right hand side is a function of the independent variable and
time only. Nystriim's method") is particularlywell-suited t o this f o r m and
gives an integration method of fourth o r d e r accuracy. The second o r d e r
s y s t e m is written as
(111-31)

(111-32)

z
-k 2 = f-(xn + -21 -n
k3 f (xn +
-

At + 1 k l (At)2 )
8 -

At + 1 k 2 (At)2)
2 -

111-23

For efficient use of computer storage a s well as computing time the com-

putations should be performed in the following o r d e r .


1. Equation (111-12) is solved using the Nystrb'm formulae (111-32). The

position of the sun and moon a r e required a t t i m e s tn, tn + 1 / 2 At,


tn + At t o be used in the evaluation of the v e c t o r s k
k
k respec-1' -2'
3
tively. It is n e c e s s a r y to p r e s e r v e the values of the vectors r
and
- EV
at these t i m e s for use in the solution of Eq. (111-27).
-EEM
2. Equations (111-27) a r e solved one s e t a t a time using formulae (111-32)

together with the values of


step.

rEvand -EM
r
which resulted f r o m the f i r s t

Many of the advantages of the r e c u r s i v e navigation method a r e now readily apparent. Although linear techniques a r e still employed, it has been possible to remove any
dependence on a r e f e r e n c e or pre-computed orbit. Within the f r a m e w o r k of a single
computational algorithm, measurement data f r o m any source may be incorporated s e quentially as obtained.

Sensitive numerical computations, such as the inversion of m a -

t r i c e s , a r e avoided.
PARAMETER ESTIMATION
The coasting flight navigation procedure just outlined is capable of generalization
to include the estimation of quantities in addition to position and velocity by the simple
expedient of increasing the dimension of the state vector beyond six. For example, one
might wish to estimate b i a s e s o r c r o s s - c o r r e l a t i o n s in the optical or r a d a r instruments,
the frequency of a satellite- borne doppler source, o r even astronomical quantities such

as distances and gravitational constants.


MEASUREMENT SCHEDULE

For effective application of this navigation method, an efficient observation schedule should be prepared. An elementary procedure, which h a s been found t o be quite effective f o r this purpose, is described in r e f e r e n c e (5). At each of a number of d i s c r e t e
times, appropriately spaced along the flight path, that measurement is selected, f r o m
a variety of possible observations, which would result in the g r e a t e s t reduction in meansquared position uncertainty a t the destination. In o r d e r to control the number of meas u r e m e n t s and prevent an unnecessarily lengthy schedule, a measurement is required
t o produce a significant reduction in the potential m i s s distance or i t will. not be made.
The simple s t r a t e g y described above, in which only currently available information is exploited, does not, of course, insure a n optimum schedule since the uncertaint i e s in position and velocity a t the t a r g e t c l e a r l y depend on the entire measurement

A method of improving a measurement schedule iteratively, employing an


adjoint of the correlation matrix, has been developed. The technique has been shown
schedule.

111-24

t o converge always t o essentially the s a m e schedule starting f r o m a variety of nominal


measurement schedules, A numerical example, reported by Denham and Speyer ( 7 ) ,
gives a minimum r m s uncertainty in position at the t e r m i n a l point which is 1 0 per cent
l e s s than the value obtained using the m o r e elementary method.

111-25

Chapter 111-3
POWER.ED-FLIGHT GUIDANCE

The t a s k of providing steering commands, frequently called guidance, s e p a r a t e s


naturally into two categories

- major

and minor maneuvers,

Launch into parking orbit,

t r a n s f e r to lunar or interplanetary orbit, insertion into orbit, and landing a r e a l l examples of major thrusting maneuvers and differ markedly f r o m the minor orbit changes
typified by mid- course velocity corrections. In either case, the guidance problem is
always a boundary value problem subject to a variety of constraints of which fuel conservation, vehicle maneuverability, and time are examples.
Explicit solutions to the problem of guidance during periods of major thrusting
r e q u i r e relatively complex calculations t o be performed in flight on a t i m e - c r i t i c a l
basis.

Considering the modest s i z e and capabilities of vehicle-borne computers, con-

t r a s t e d with the m o r e familiar commercial machines, the design of feasible explicit


methods p r e s e n t s a considerable challenge. Several of the m o r e promising guidance
techniques currently under development a r e compared in this chapter.
ADAPTIVE GUIDANCE MODE
The guidance method developed at Marshall Space Flight Center f o r the Saturn
rocket and t e r m e d the Adaptive Guidance Mode(*) is conceptually simple and easily
described.

The f-o r m of the guidance and cut-off equations is invariant with changing

missions and vehicles and, therefore, is in accord with the requirements of a unified
method. However, significantly l a r g e quantities of ground computations a r e required
to determine certain coefficients needed in the mechanization.
The vector values of position and velocity, the s c a l a r magnitude of thrust acceleration and t i m e a r e updated continuously during powered flight, At each instant the
present values of these quantities may be considered as initial conditions for the r e mainder of the flight.

Ideally, one would determine the optimum t r a j e c t o r y f r o m p r e -

sent conditions t o desired t e r m i n a l conditions and command a thrust direction f r o m this


optimum solution. This is, of course, impractical so that the techniques of the calculus

of variations a r e employed to generate a volume of expected trajectories for specific


vehicles and missions. Numerical curve fitting methods a r e employed to obtain satisfactory s e r i e s solutions for the guidance and cut-off commands.

111-27

A functional diagram for the Saturn guidance system is shown in Fig. 111-6. D u r ing flight the thrust acceleration magnitude is computed approximately once per second
by differentiating the outputs of the integrating a c c e l e r o m e t e r s and taking the square
root of the sum of the s q u a r e s of the resulting derivatives.

Guidance and cut-off com-

mands a r e computed as polynomial functions of position, velocity, thrust acceleration


and time a t intervals of approximately one second.

During the burning of the f i r s t stage

of Saturn, the guidance program is obtained as a polynomial expansion in time only because of s t r u c t u r a l and control problems.
The chief difficulty with the Adaptive Guidance Mode is determining the best method of representing the volume of expected trajectories which provide minimum fuel consumption.

The required number of t e r m s in the polynomials to obtain acceptable a c -

curacy has been found t o v a r y f r o m 40 to 6 0 depending on the mission.


VELOCITY-TO-BE-GAINED METHODS
Conic orbits can be exploited t o advantage in solving many guidance problems.
F o r those major orbital t r a n s f e r maneuvers which can be accomplished conceptually by

a single impulsive velocity change, an instantaneous velocity-to-be-gained vector based


on conic orbits can often be defined and the vehicle s t e e r e d to null this vector.
Refer to Fig. 111-7 and let a vector v

-r be defined, corresponding to the present

vehicle location r, as the instantaneous velocity required to satisfy a s e t of stated m i s sion objectives. The velocity difference v between
-g
v is then the instantaneous velocity-to-be-gained,

vr and the present vehicle velocity

Two convenient guidance l a w s a r e immediately apparent which will a s s u r e that all


t h r e e components of the vector v a r e simultaneously driven to zero. F i r s t , we may
-g
orient the vehicle t o align the thrust acceleration vector
with the direction of the
velocity-to-be-gained vector.

Alternatively, since a convenient expression can be de-

veloped for the time r a t e of change of the v vector, we may direct the vector
to
-g
cause the vector
t o be parallel to v and oppositely directed. If the thrust a c c e l e r -g
-g
ation magnitude is not sufficiently l a r g e it may not be possible to align the vector v
with i t s derivative.

However, with typical chemical rockets for which the burning-g


time

is relatively short, no difficulty has been encountered with this guidance logic,
A combination of these two techniques leads t o a highly efficient s t e e r i n g law which
compares favorably with calculus of variations optimum s o l u t i ~ n s ! ~ ) T hsec a l a r mixing
p a r a m e t e r y is chosen empirically to maximize fuel economy during the maneuver. A
constant value of y is usually sufficient for a particular mission phase; however, if r e 4

quired, it may be allowed t o v a r y as a function of some convenient system variable.


t

I11 -2 8

UJ

Lu

-da
0
Lu

>

i=

3a

I11- 2 9

>-

VI
k

VI
0

Ei
c?
I

Ql

II

.. ..

.. .. ..

I1

t=l

0
I1

111- 3 0

A functional diagram illustrating the computation of the e r r o r signal required for


control purposes is shown in Fig. 111-8.

The position, velocity and gravitation vectors

a r e computed f r o m the outputs of integrating a c c e l e r o m e t e r s as described e a r l i e r in the


section on navigation.

The required impulsive velocity needed to achieve mission ob-

jectives is determined as a function of the position vector and used to calculate the
velocity-to-be-gained.

Numerical differentiation of the required velocity vector and the

a c c e l e r o m e t e r outputs, using values stored f r o m the previous sample time, prsvides


two important ingredients of the e r r o r signals. When properly scaled, the system output is a vector r a t e of command whose magnitude is proportional t o the s m a l l angular
differences between the actual and commanded thrust acceleration v e c t o r s and whose
direction defines the direction of vehicle rotation required to null the e r r o r .

Near the

end of the maneuver, when the velocity-to-be-gained is small, cross- product steering

is terminated, the vehicle holds a constant attitude and engine cut-off is made on the
b a s i s of the magnitude of the v vector.
-g
This guidance technique is being considered for s t e e r i n g the Apollo Command
Module during the following mission phases: (1) translunar injection which r e f e r s to
the p r o c e s s of t r a n s f e r f r o m e a r t h parking orbit to a t r a j e c t o r y linking e a r t h and moon;
(2) t r a n s f e r f r o m a hyperbolic approach t r a j e c t o r y t o a c i r c u l a r orbit of the moon; and
(3) t r a n s e a r t h injection or t r a n s f e r f r o m a lunar orbit to an earth-bound trajectory.
For each of these maneuvers, the required impulsive velocity is a s follows:
A.

T r a n s l u n a r Injection
The required velocity f o r translunar injection is defined as that velocity,

at the present position, that w i l l place the vehicle on a conic passing through
a specified time.

Specifically, this velocity vector

xr is calculated f r o m

(111-33)

where

sgn(tm - t )

1
JL--

2a

In t h e s e formulae, c is the linear distance f r o m the present position r t o the


s is the s e m i - p e r i m e t e r of the triangle formed by the
t a r g e t position
r and r T ; a is the semimajor axis of the conic; t is the t i m e of flight:
vectors and t m is the time to fly the minimum energy path f r o m r- t o -T
r
The choice

rT:

I11-31

VECTOR
CROSS
PRODUCT

NAVIGATION
COMPUTATIONS

VELOCITY -TO-BE-GAINED

INERTIAL

v-g

't

INERTIAL
POSITION

CDh\llTV

A
INTEGRATING
ACCELEROMETER
OUTPUT

SENSOR
SYSTEM

H
H

w
N

(tn)

t'

Xa(tn-11

l-I

'

INCREMENTAL
COMMANDED
ROTATION RATE

AE

L
Fig. 111-8

Velocity-to-beGained Steering

of upper o r lower sign in the expression f o r A is made according as the


t r a n s f e r angle is l e s s than or g r e a t e r than 180, respectively.
The t a r g e t point is actually offset by a calibrated amount f r o m the d e s i r e d position t o account f o r gravitational perturbations. T o simplify the
computational load the fixed t i m e requirement is readily approximated by
holding constant the semimajor a x i s of the conic at a pre-determined value.
B.

C i r c u l a r Orbit Insertion
T o guide a vehicle into a c i r c u l a r orbit of the moon by a rocket braking
maneuver initiated on a n approach trajectory, the vector v may be de-r
fined as that velocity impulse required at the present position to c i r c u l a r i z e
the orbit in a specified plane.
lative t o the moon and i

"n

If r is the position vector of the vehicle r e -

is the unit normal t o the d e s i r e d orbit plane, then

(Ill- 34)

The shape and orientation of the final orbit is controlled by this means
but d i r e c t control of the orbital r a d i u s is not possible.

However, t h e r e is

a n e m p i r i c a l relationship between the final r a d i u s and the pericenter of the


approach trajectory, so that a d e s i r e d r a d i u s can be established by a n appropriate selection of the approach orbit.
C.

T r a n s e a r t h Injection
In the vicinity of the moon the spacecraft t r a j e c t o r y is v e r y n e a r l y hyperbolic.

Therefore, the required velocity f o r t r a n s e a r t h injection f r o m lunar

orbit may be conveniently defined by the magnitude v, and direction i, of


the asympotic velocity 1,.

Thus
(111-35)

The direction in space that the t h r u s t vector should be oriented a t the beginning of a powered flight maneuver is determined f r o m the equation

where i is a unit vector in the direction of the v vector and


-g
-g

111-33

The quantities v and p a r e both continuous functions through the ignition


-g
point and, thus, their computation can be s t a r t e d t o align the vehicle initially
prior to the firing of the engine.
TER.MINAL STATE VECTOR. CONTROL
The explicit technique just discussed is workable if it is possible, at thrust t e r m i nation, to define the required velocity as a function of position and thereby eliminate
the need f o r position control. On the other hand, when burn-out position and velocity
a r e independently specified, a n alternate guidance method, based on a n explicit solution
of the powered flight dynamics, is frequently applicable(lO). As examples, consider
the problems of insertion of a vehicle into a c i r c u l a r orbit a t a pre-specified altitude
which lies in a prescribed plane, soft-landing a vehicle on the surface of the moon and
orbital rendezvous.
The guidance computations, needed f o r solving explicitly the m o r e general bounda r y value problem, involve a determination of the time remaining before thrust t e r m i nation. For fixed t h r u s t rocket problems, the termination time is calculated cyclically
by a n iteration p r o c e s s in such a manner a s to control the final velocity along one coordinate axis. A s a part of the calculation, the effective exhaust velocity of the rocket
engine, based on a mathematical model of the engine performance, is needed.
When the vehicle is propelled by a controllable thrust engine, the magnitude of
the t h r u s t acceleration can be commanded t o cause burn-out t o occur at a pre-specified
t e r m i n a l time. In t h i s case, the time-to-go is a trivial calculation. P r i o r to thrust
initiation, the thrust termination time is chosen according t o c r i t e r i a which depend on
the particular guidance problem. F o r orbital rendezvous, the time and desired t e r m i nal position and velocity a r e chosen f r o m a knowledge of the target vehicle ephemeris,
F o r a lunar landing, the t e r m i n a l time is selected t o maximize the initial thrust accel-

eration.
The development of a n explicit s t e e r i n g equation f o r a controllable thrust engine,
which will guide a vehicle to a desired s e t of terminal conditions, is based on the solution t o the following simple variational problem. Let it be required t o find the acceleration program a(t), which will minimize the functional

(111-37)

111-34

where t is present time and t D is the desired terminal time.

If a (t) is the tota.1 a c c e l -

eration influencing the vehicle motion, then


dr

dv
- v

(111-38)

"

dt
subject to the boundary conditions

(111-39)

This minimization problem is readily solved using the calculus of variations.

X and
introducing the two vector Lagrange multipliers -

-r ) we may combine Eqs.

By

(111-37)

and (111-38) in the f o r m

for which the Euler- Lagrange equations a r e found to be

-dtd x -T --

(111-40)

The solution of Eqs. (111-40) yields

a(t) =

c1

and the constants of integration


and
(111-39). The final result is simply

c 1+c2t

c z a r e chosen to satisfy the boundary conditions

where
t

go

=tD-t

is the time- to- go before termination.

I11-35

-r

111- 3 6

In a guidance maneuver the total acceleration a(t) is the sum of thrust acceleration

a .(t)

and gravity g ( r ) . If the gravity vector were a constant, then the exact solution to
the guidance problem would be
"

E T = t
go

(VD

- v, '7 [ X D - (5

XD tgO)l

-g

(111-41)

go

In problems of practicalinterest, the vector g is not constant and the integrals q u a r e criterion of Eq. (111-37) is not appropriate for fuel minimization.

However, it

happens that Eq. (111-41) does provide a nearly optimum s t e e r law for a wide variety of
problems.
F i g u r e 111-9 illustrates the guidance computations required to achieve a given
t e r m i n a l position and velocity a t a specified time with a throttleable rocket engine.
F r o m the navigation s y s t e m the p r e s e n t position, velocity, gravitational acceleration
and time a r e determined and the direction and magnitude of the thrust acceleration to
be commanded a r e calculated. A s the t e r m i n a l conditions a r e approached, time-to-go
approaches z e r o and the computation c l e a r l y becomes unstable. The singularity is
readily avoided, with only slight loss in potential performance, by holding time-to-go
constant in the guidance expression when it is l e s s than a pre- assigned amount. Engine
cut-off can then be commanded when the actual time-to-go r e a c h e s zero.

.
c

I11- 3 7

Chapter 111-4
MID-COURSE GUIDANCE

LINEAR.IZED GUIDANCE THEORY


Techniques of guidance and navigation of a spacecraft in interplanetary or cislunar
space a r e often based on the method of linearized perturbations.

The approach is t o lin-

e a r i z e the equations of motion by a s e r i e s expansion about a nominal o r r e f e r e n c e orbit


in which only f i r s t - o r d e r t e r m s a r e retained.

The resulting equations a r e far s i m p l e r

and superposition techniques, as well as all of the powerful tools of linear analysis,
may be exploited to obtain solutions t o a wide variety of navigation and guidance problem.
Consider, for example, the guidance problem illustrated in Fig. 111-10. A vehicle

one

is launched into orbit a t time t

and moves under the influence of


or m o r e gravity
L
fields t o r e a c h a target point a t t i m e tA. Let r o ( t n ) and x o ( t n ) be the position and velocity vectors a t time tn for a vehicle traveling along a r e f e r e n c e path connecting the
r(t )
initial and final points. Because of e r r o r s , the t r u e position and velocity v e c t o r s "
n

and v ( t n ) will deviate f r o m the associated r e f e r e n c e quantities.

If the deviations f r o m

the r e f e r e n c e path a r e always small, so that linearization techniques are applicable, the
velocity correction Av ':< may be computed as a linear combination of the position and
-n
velocity deviations. The three-dimensional m a t r i x C * 'is the m a t r i x of partial derivn .
atives, with r e s p e c t t o the components of r, of the components of the velocity vector v*

required to r e a c h the target f r o m position r.

For these calculations t o r e m a i n valid it i s , of course, n e c e s s a r y t o r e s t r i c t the


magnitude of the deviations f r o m the corresponding nominal values.

Another disadvan-

tage of the method is that all possible t i m e s of velocity corrections must be anticipated
and associated values of the C* matrix stored in the guidance computer. Also to provide
an adequate launch window, a family of r e f e r e n c e t r a j e c t o r i e s is mandatory and the guidance computer storage r e q u i r e m e n t s rapidly become excessive using this approach.
EXPLICIT TECHNIQUES
The quantity of stored data required for mid- course guidance maneuvers can be
markedly reduced if explicit techniques a r e employed using conic a r c s suitably modified

111-39

>I

a
I

C
L I

40
* C

U
II

'1 c
*I

"

111-40

to account for s m a l l non-central force field effects.

Both fixed and variable -time-of-

a r r i v a l velocity corrections can be calculated and the procedures w i l l be illustrated by


two specific examples.
Fixed-Time-of-Arrival Guidance

Because of initial e r r o r s a r i s i n g f r o m a failure

to inject the spacecraft in an appropriate t r a j e c t o r y to the moon, a velocity correction


is frequently required after a few hours of coasting flight.

During the post-injection

phase, an a c c u r a t e determination of the vehicle's orbit is made using navigation t e c h An intermediate target point r

niques as previously discussed.

-T

is selected as the

position on the lunar s p h e r e of influence through which the r e f e r e n c e vehicle would p a s s


at the r e f e r e n c e time.

Refer to Fig. 111-11 and let r and v be the position and velocity estimates of the
vehicle at the t i m e a correction is to be made.

Using the t r a j e c t o r y integration routine,

which is a p a r t of the coast phase navigation program, the position of the vehicle is extrapolated to determine the point
at which the spacecraft would be found at the t a r get r e f e r e n c e time if no c o r r e c t i v e action w e r e taken. By calculating the conic a r c

connecting the position vectors r and -T


r
in the s a m e time interval, the conic velocity
vector v
at r is determined. The difference between the conic velocity and the vehi-cl
cle's actual velocity is a good m e a s u r e of the effect of lunar and s o l a r perturbations.

A second conic a r c connecting the vehicle position vector r and the desired target point

r produces the conic velocity vector -22'


v
If this velocity is c o r r e c t e d for the effect
-T
of perturbations, the velocity n e c e s s a r y to r e a c h the desired target f r o m position r- is
obtained. Thus, an excellent approximation to the required velocity correction is just
the difference between the two conic velocities. The computation may, of course, be
repeated iteratively to achieve any d e s i r e d degree of convergence.

However, in p r a c -

tice, one computation cycle is usually sufficient.


Fixed- time- of- arrival guidance may be summarized as follows:

A. The conic velocity required at r to a r r i v e a t -T


r

-c l =
v

sgn ( r 2 x)

'ic

-ir)
f

is calculated f r o m

- YC(Y)I
( i c + -r
i )
4(s - c)

(111-42

where pE is the gravitational constant of the earth, 2, is a unit vector in the


r, and s is the s e m i direction of r, i is a unit vector in the direction of
"

perimeter

of the triangle formed by r and

determined as the r o o t s of the equations

LII -41

rT'- -

rT'. The quantities x and y a r e

zii

+I-

111-42

(111-43)

where At is the time difference between the r e f e r e n c e time of a r r i v a l and


present time.

The special transcendental functions S and C are defined

in Eqs. (111-11). The choice of the upper and lower signs in E q s . (111-42)
and (111-43) is made according to whether the angle between r and r

is

l e s s than or g r e a t e r than 180 degrees, respectively.

B.

C.

The conic velocity v


for attaining r is computed by repeating step A
"c2
-T
with r substituted f o r
-T
The estimated velocity correction is then given by

xT1.

Variable- Time- of- Arrival Guidance


When a velocity correction is made i n the
vicinity of the moon, the arrival time a t perilune may be allowed t o v a r y thereby r e ducing substantially the required velocity correction as well as the t e r m i n a l velocity
deviation f r o m i t s nominal value.

Specifically, let the desired terminal conditions at

the moon be a specified altitude a t perilune and a fixed plane in which the perilune vect o r is t o lie.

Again, as shown in Fig. 111-12, the t r a j e c t o r y is extrapolated forward in

t i m e t o locate the perilune vector r ' which would result in the absence of a velocity
-P
correction, A conic a r c with r ' as perilune and connecting the position vector r is
-P
then determined to obtain a m e a s u r e of the gravitational perturbation. The desired

perilune vector r is calculated f r o m r


by scaling i t s length to correspond to the r e -P
-P
quired perilune distance and then rotating it into the required plane while keeping the
c e n t r a l angle f3 fixed.

A second conic a r c with r as perilune is calculated and the


-P
difference between the two conic velocities again provides an excellent approximation
to the n e c e s s a r y velocity correction.
Theoretically, the desired plane should not be fixed in space, but should rotate
with the moon.

However, the change i n perilune a r r i v a l time combined with the moon's

own rotation l e a d s to t e r m i n a l deviations which a r e s m a l l e r than the navigation u n c e r tainties.

Hence, it is sufficiently a c c u r a t e t o a i m f o r a fixed plane when approaching

perilune.

111-43

p.

In -44

Perilune guidance may be summarized as follows:

A.

The conic velocity v c l required at r to attain pericenter at r


-P
puted from

is corn-

where 6 is the angle between r and r I and p, the p a r a m e t e r of the conic,


-P '
is given by

rr
P=
B.

(1

- cos e)

(111-46)

-rcos8
P
is rotated into the desired plane and scaled to

The pericenter vector r


-P
the desired length r by means of
P

-P

= r

[ d l + pcos8'

U N I T-n( i X-i r ) + p i

"n

X(i

"n

(111-47)

X i )]

-r

where

is normal t o the d e s i r e d plane in the direction of the


The unit vector in
angular momentum vector.

c.

The conic velocity c2, required to attain pericenter at r


is then calP'
culated by repeating step A with r in place of r l .

"p

D.

"p

The magnitude of the required velocity correction may be f u r t h e r r e duced be noting that t h e r e is a direction along which a velocity change
may be made without altering the altitude a t pericenter.

If the component

of velocity correction along this insensitive direction is deducted from the


total correction, the effect will be simply a s m a l l rotation of the perilune
vector r
-P'
"D
i

This insensitive direction is computed f r o m

UNIT[- (1

- cos

"
i3:

+ s i n 8 1 - cos 8 +

is a unit vector in the direction of r

whereir
P

$)ir
X i
-n

-P'

(111-48)

111-46

The estimated velocity correction is then given by the component of the vec-

E.

tor v c2

-I

C 1

in the plane perpendicular t o i D and is calculated f r o m

During t r a n s e a r t h flight it is not sufficient to a i m f o r a fixed plane when making a


velocity correction to the vicinity of the earth.

The d e s i r e d t e r m i n a l conditions a r e a

vacuum perigee distance (which is equivalent to a n entry angle) and a landing s i t e fixed
to the earth.

This type of velocity correction, called perigee guidance, is an extension

of perilune guidance with the plane determined so that the spacecraft w i l l be directed to
the d e s i r e d landing site.
Inherent in perigee guidance is a timing problem which n e c e s s i t a t e s taking into
account the correction to be made in estimating the t i m e of a r r i v a l at perigee. The
change in perigee time due to a correction that a l t e r s the perigee distance f r o m r I to
P
r is given by the emperically determined formula k r (r
r ), where k has been found
P
P
P
experimentally t o be 1 6 X 1 0 - l ' h r / m i 2 . A simple calculation shows that a velocity
correction, which is made at the lunar s p h e r e of influence (about 200, 000 m i l e s f r o m

the earth) and which changes the perigee distance by 500 miles, a l t e r s the perigee a r -

~*

r i v a l time by nearly ten minutes.


Let St be the estimated deviation in perigee a r r i v a l t i m e in hours, i
a unit
P
LO
vector in the direction of the landing site a t the nominal time of arrival, and cu0 the

nominal angle from perigee to i


. Assuming that the spacecraft travels on the average
-LO
a t c i r c u l a r orbital speed during entry, the deviation in the angle through which the e a r t h
rotates is given by
(111-50)
where CY is the actual angle f r o m perigee to the landing site. When the e a r t h r o t a t e s
to i according to
through an angle 6A, the landing site changes f r o m i
-Lo
-L

/ cos 6A

- sin 6A

O \

(111-51)
0

The angle

CY

satisfies
cos -1 (i

-r

-i L ) - 0
(111-52)

2a

- cos -1 (ir' iL)


-e

111-47

where the choice of the f i r s t o r second equation is made according to whether the angle
b e t w e e n i r and i

is g r e a t e r or l e s s than 1 8 0 d e g r e e s , respectively. The desired plane

"

is then determined by the initial position r and the landing site vector -iL calculated from

E q s . (111-50)., (111-51) and (111-52).


If the spacecraft t r a j e c t o r y was, indeed planar, the s t e p s outlined in the discussion of perilune guidance would be adequate f o r calculating the velocity correction.
Unfortunately, the non-planar c h a r a c t e r i s t i c s a r e sufficiently pronounced that an additional s t e p is required before the perilune guidance technique can be applied.
In Fig. 111-13 is represented, schematically, and edge-wise view of the t r a j e c t o r y
problem in which a planar path a p p e a r s as a straight line.

The unmodified perilune

guidance method would cause the vehicle to head f o r a landing site a t -iL
To counteract this effect, the vector r

I'

instead of i L'

iL',
the

position the
-P
s p a c e c r a f t would achieve on a n uncorrected t r a j e c t o r y at the t i m e the t a r g e t landing site

is

is projected ahead to

Then a false perigee position r

I'

"p

'

i n the plane determined by r and iL1,


is

used in place of r I .
-P
Perigee guidance m a y be summarized as follows:
A.

The estimated change in t i m e of a r r i v a l a t perigee is calculated from


6t

6t

+ kr (r - r
P
P

(111-53)

I is the deviation i n perigee time obtained in the extrapolation of


P
r and v t o perigee r ' ,
-P
The position of the d e s i r e d landing site is found by solving the transcendental

where 6 t

B.

equations (111-50), (111-51) and (111-52) f o r i L and


C.

cy.

The unit v e c t o r i L ' is calculated f r o m


(111-54)
where i
-r

D.

and i
a r e unit v e c t o r s i n the directions of the position and velo-V
P
P
city v e c t o r s at r I .
-P
The f a l s e perigee vector is located using Eq. (111-47) with the substitutions

I11-4 8

E.

The unit vector normal to the desired plane is computed f r o m


i

"n

F.

*UNIT(ir X iL)
"

The remaining calculations a r e exactly as outlined in steps A through E f o r


perilune guidance.

in s t e p s D and
In the determination of the vector in

E above, the upper or lower sign is selected according to whether the angle
between i and either i L ' o r i
whichever is relevant, is g r e a t e r o r l e s s
-r
-E
than 180 degrees, respectively.

OPTIMUM GUIDANCE POLICIES


In an effort to compensate for initial e r r o r s by means of a mid- course velocity
correction, new e r r o r s will inevitably be made which must again be corrected.

The

problem of determining when and how t o perform impulsive c o r r e c t i o n s to a spacecraft


orbit can be classified as a multistage decision process. Various guidance policies
have been proposed and the new mathematical techniques of dynamic programming and
steepest ascent optimization theory have been used with some s u c c e s s in an attempt to
formulate an optimum policy. Several useful guidance policies a r e described and compared by Curkendall and Pfeiffer'")

in a r e c e n t paper.

A s with all applications of dynamic programming techniques, the computational


r e q u i r e m e n t s a r e extensive and rapidly become impractical as the dimension of the problem i n c r e a s e s .

The r e s u l t s obtained by Arcon(12) and Orford(13) using this approach

have been r a t h e r limited and numerical examples a r e r e s t r i c t e d t o problems of only


one or two dimensions r a t h e r than six.
Denham and S ~ e y e r ' ~have
)
formulated a method of improving a velocity c o r r e c tion schedule iteratively using steepest ascent techniques in a manner s i m i l a r to that
mentioned e a r l i e r for optimizing measurement schedules.

No numerical r e s u l t s a r e

yet available.
Unfortunately, the Monte Carlo approach t o the determination of velocity- correction t i m e s r e m a i n s the most practical method.

The lack of mathematical elegance and

an over-abundance of computer usage time, which c h a r a c t e r i z e this technique, a r e ,


nevertheless, balanced by the capability of utilizing a realistic, r a t h e r than an o v e r simplified, mathematical model.

111-4 9

BIBLIOGR.APHY

1.

Bumstead, R.obert M. and Wallace E. Vander Velde: Navigation and


Guidance Systems Employing a Gimballess IMU, P r o g r e s s in Astronautics and Aeronautics, vol. 13, pp. 391-419, Academic P r e s s , New

York, 1964.
2.

Wiener, Thomas F. : Theoretical Analysis of Gimballess Inertial


Reference Equipment Using Delta-Modulated Instruments, MIT
Instrumentation Laboratory ScD T h e s i s T-300, Cambridge, M a s s . ,
March 1962.

3.

Blackman, R . B. : Methods of Orbit Refinement, Bell Telephone


Technical Journal, vol. 43, pp. 885-909, May, 1964.

4,

Potter, James E. and R.obert G. Stern: Statistical Filtering of Space


Navigation Measurements, P r o g r e s s in Astronautics and Aeronautics,
vol. 13, pp. 775-801, Academic P r e s s , New York, 1964.

5.

Battin, R.ichard H. : "Astronautical Guidance", McGraw-Hill Book


Company, Inc., New York, 1964.

6.

Henrici, Peter: " Discrete Variable Methods in Ordinary Differential


Equations", John Wiley and Sons, Inc. , New York, 1 9 6 2 .

7.

Denham, Walter F. and J a s o n L. Speyer: Optimal Measurement and


Velocity Correction P r o g r a m s f o r Midcourse Guidance, AIAA Journal,
vol. 2 , pp. 896-907, May, 1964.

8.

Moore, F . Brooks and Melvin Brooks: Saturn Ascending Phase Guidance and Control Techniques, P r o g r e s s in Astronautics and Aeronautics, vol. 10, pp, 183-209, Academic P r e s s , New York, 1963.

9.

Martin, F r e d e r i c k H. : Closed-Loop Near-Optimum Steering for a


C l a s s of Space Missions, MIT Instrumentation Laboratory ScD Thesis
T-413, Cambridge, Massachusetts, May, 1965.

10.

Cherry, George W. : E Guidance: A Universal, Explicit, Optimizing


Guidance Law for Rocket- Propelled Flight, AIAA/ ION Astrodynamics,
Guidance and Control Conference P a p e r 64-638, Massachusetts
Institute of Technology, Cambridge, Massachusetts, August, 1964,

I11 51

11.

Curkendall, D. W . and C. G. Pfeiffer: Discussion of Guidance Policies


for Multiple-Impulse Corrections of the T r a j e c t o r y of a Spacecraft,

P r o g r e s s in Astronautics and Aeronautics, vol. 13, pp. 667-687,


Academic P r e s s , New York, 1964.
12.

Advanced R.esearch Consultants: Some Aspects of Midcourse Guidance,


Report R 6 2 - 5 , Lexington, Massachusetts, July, 1962.

13.

Orford, Richard J. : Optimization of Stochastic F i n a l Value Control


Systems Subject t o Constraints, Grumman Aircraft Engineering
Corporation, R.eport R.E-157, Bethpage, New York, May, 1962.

8 1

111-52