Spacecraft and Aircraft Dynamics

Matthew M. Peet
Illinois Institute of Technology

Lecture 10: Linearized Equations of Motion

Aircraft Dynamics
Lecture 10

In this Lecture we will cover: Linearization of 6DOF EOM Linearization of Motion Linearization of Forces
Discussion of Coecients

Longitudinal and Lateral Dynamics Omit Negligible Terms

Decouple Equations of Motion

M. Peet

Lecture 10:

2 / 19

Review: 6DOF EOM

Fx u + qw rv Fy = m v + ru pw Fz w + pv qu and L Ixx p Ixz r qpIxz + qrIzz rqIyy M = Iyy q + p2 Ixz prIzz + rpIxx r2 Ixz N Ixz p + Izz r + pqIyy qpIxx + qrIxz

M. Peet

Lecture 10:

3 / 19

Linearization
Consider our collection of variables: X, Y, Z, p, q, r, L, M, N, u, v, w... To Linearize: Step 1: Choose Equilibrium point: X0 , Y0 , Z0 , p0 , q0 , r0 , L0 , M0 , N0 , u0 , v0 , w0 . Step 2: Substitute. u(t) = u0 + u(t) p(t) = p0 + p(t) X (t) = X0 + X (t) L(t) = L0 + L(t) v (t) = v0 + v (t) q (t) = q0 + q (t) Y (t) = Y0 + Y (t) M (t) = M0 + M (t) w(t) = u0 + w(t) r(t) = r0 + r(t) Z (t) = Z0 + Z (t) N (t) = N0 + N (t) also dont forget , , .

Step 3: Eliminate small nonlinear terms. e.g. u(t)2 = 0,

M. Peet

u(t)r(t) = 0,
Lecture 10:

etc.
4 / 19

Linearization
Step 1: Choose Equilibrium

For steady ight, let u0 = 0 p0 = 0 X0 = 0 L0 = 0 v0 = 0 q0 = 0 Y0 = 0 M0 = 0 w0 = 0 r0 = 0 Z0 = 0 N0 = 0

Other Equilibrium Factors

Weight Weight: Pitching Motion changes force distribution.

X0 = mg sin 0 Z0 = mg cos 0
M. Peet Lecture 10: 5 / 19

Linearization
Step 2: Substitute into EOM

We use trig identities and small angle approximations ( small): sin(0 + ) = sin 0 cos + cos 0 sin = sin 0 + cos 0

cos(0 + ) = cos 0 cos sin 0 sin = cos 0 sin 0

M. Peet

Lecture 10:

6 / 19

Linearization
Step 2: Substitute into EOM

Substituting into EOM, and ignoring 2nd order terms, we get X m Y v g cos 0 + u0 r = m Z w + g sin 0 u0 q = m u + g cos 0 = p = Ixz Izz L + N 2 2 Ixx Izz Ixz Ixx Izz Ixz M q = Iyy Ixz Ixx r = L + N 2 2 Ixx Izz Ixz Ixx Izz Ixz

Note these are coupled with , .

M. Peet Lecture 10: 7 / 19

Linearization
Step 2: Substitute into EOM

We include expressions for , . = q = p + r tan 0 = r sec 0 For steady-level ight, 0 = 0, so we can simplify = q = p = r which is what we will mostly do.

M. Peet

Lecture 10:

8 / 19

Linearization
Step 2: Substitute into EOM

We can also express the equations for translational motion x = u cos 0 u0 sin 0 + w sin 0 y = u0 cos 0 + v z = u sin 0 u0 cos 0 + w cos 0

So now we have 12 equations and 12 variables. But Wait But Wait!!! Theres More!!! Recall the forces and moments depend on motion and controls: e.g. X (u, v, w, , e , t ).
More Variables! More Nonlinear Terms!

M. Peet

Lecture 10:

9 / 19

Linearization
Force Contribution

Out of (u, v, w, u, v, w, p, q, r, a , e , r , T ), we make the restrictive assumptions on form (Why?): X (u, w, e , T )

Y (v, p, r, r ) Z (u, w, w, q, e , T ) L(v, p, r, r , a ) M (u, w, w, q, e , T ) N (v, p, r, r , a )

where we have following new variables T - Throttle control input.

e - Elevator control input. a - Aileron control input. r - Rudder control input.

It could be worse (, , ). Reality is worse.

M. Peet Lecture 10: 10 / 19

Linearization
Force Contribution

To linearize the forces/moments we use rst-order derivative approximations: X = Y = Z = L = M = N = X X X X e + T u + w + u w e T Y Y Y Y r v + p + r + v p r r Z Z Z Z Z Z e + T u + w + w + q + u w w q e T L L L L L r + a v + p + r + v p r r a M M M M M M e + T u + w + w + q + u w w q e T N N N N N r + a v + p + r + v p r r a

M. Peet

Lecture 10:

11 / 19

Linearization
Coecients

We have a notation for the partial derivatives:

1 X m u 1 Y m v 1 Z m u 1 L Izz v 1 M Iyy u 1 N Ixx v = Xu = Yv = Zu = Lv = Mu = Nv 1 X m w 1 Y m p 1 Z m w 1 L Izz p 1 M Iyy w 1 N Ixx p = Xw = Yp = Zw = Lp = Mw = Np 1 X m e 1 Y m r 1 Z m w 1 L Izz r 1 M Iyy w 1 N Ixx r = Xe = Yr = Zw = Lr = Mw = Nr 1 X m T 1 Y m r 1 Z m q 1 L Izz r 1 M Iyy q 1 N Ixx r = XT = Yr = Zq = Lr = Mq = Nr 1 Z m e 1 L Izz a 1 M Iyy e 1 N Ixx a = Ze = La = Me = Na

M. Peet

Lecture 10:

12 / 19

Longitudinal Dynamics
Although we now have many equations, we notice that some of them decouple: X m Z w + g sin 0 u0 q = m M q = Iyy = q u + g cos 0 = where 1 X = Xu u + Xw w + Xe e + XT T m 1 Z Z e + ZT T Z = Zu u + Zw w + w + Zq q + m w e 1 M = Mu u + Mw w + Mw + Mq q + Me e + MT T w Iyy and also, x = u cos 0 u0 sin 0 + w sin 0 z = u sin 0 u0 cos 0 + w cos 0
M. Peet Lecture 10: 13 / 19

Simplied Longitudinal Dynamics

= q Note X m = Z w + g sin 0 u0 m = M Iyy u + g cos 0 =

where 1 X = Xu u + Xw w + Xe e + XT T m 1 + Z e + Z T Z = Zu u + Zw w + Zw + Zq w T e m 1 + M e + M T M = Mu u + Mw w + Mw + Mq w T e Iyy
M. Peet Lecture 10: 14 / 19

Simplied Longitudinal Dynamics

Combining:
u + g cos 0 = Xu u + Xw w + Xe e + XT T = Zu u + Zw w + Zw + Ze e + Z T w + g sin 0 u0 + Zq w T = Mu u + Mw w + Mw + Mq + Me e + MT T w

Homework: Put in state-space form. Hint: Watch out for w .

M. Peet

Lecture 10:

15 / 19

Lateral Dynamics
The rest of the equations are also decoupled: v g cos 0 + u0 r = p = Y m

Izz Ixz L + N 2 2 Ixx Izz Ixz Ixx Izz Ixz Ixx Ixz L + N r = 2 2 Ixx Izz Ixz Ixx Izz Ixz = p + r tan 0 where 1 Y = Yv v + Yp p + Yr r + Yr r m 1 L = Lv v + Lp p + Lr r + Lr r + La a Izz 1 N = Nv v + Np p + Nr r + Nr r + Na a Ixx y = u0 cos 0 + v
Lecture 10: 16 / 19

and also
M. Peet

Alternative Representation

M. Peet

Lecture 10:

17 / 19

Conclusion

Today you learned Linearized Equations of Motion:

Linearized Rotational Dynamics Linearized Force Contributions New Force Coecients

Equations of Motion Decouple

Longitudinal Dynamics Lateral Dynamics

M. Peet

Lecture 10:

18 / 19

Conclusion

Next class we will cover: Longitudinal Dynamics: Finding dimensional Coecients from non-dimensional coecients Approximate modal behaviour
short period mode phugoid mode

M. Peet

Lecture 10:

19 / 19