Feedback linearization

Control mapping

Nonlinear plant

u = a(x) + b(x)v
that will render a linear inputoutput map from the
new input v R to the output y . To ensure that
the transformed system is an equivalent representation
of the original system, the transformation must be a
dieomorphism. That is, the transformation must not
only be invertible (i.e., bijective), but both the transformation and its inverse must be smooth so that dierentiability in the original coordinate system is preserved in the
new coordinate system. In practice, the transformation
can be only locally dieomorphic, but the linearization
results only hold in this smaller region.

Linear plant
Linear controller

Block diagram illustrating the feedback linearization of a nonlinear system

We require several tools before we can solve this problem.

Feedback linearization is a common approach used in
controlling nonlinear systems. The approach involves
coming up with a transformation of the nonlinear system 1.1 Lie derivative
into an equivalent linear system through a change of variables and a suitable control input. Feedback linearization The goal of feedback linearization is to produce a transformed system whose states are the output y and its rst
may be applied to nonlinear systems of the form
(n 1) derivatives. To understand the structure of this
target system, we use the Lie derivative. Consider the
time derivative of (2), which we can compute using the
x = f (x) + g(x)u
(1)
chain rule,
y = h(x)
(2)
where x Rn is the state vector, u Rp is the vector of
inputs, and y Rm is the vector of outputs. The goal is
to develop a control input

y =

u = a(x) + b(x)v

Now we can dene the Lie derivative of h(x) along f (x)

as,

d h(x)
d h(x)
=
x
dt
dx
d h(x)
d h(x)
=
f (x) +
g(x)u
dx
dx

that renders a linear inputoutput map between the new

input v and the output. An outer-loop control strategy for
d h(x)
Lf h(x) =
f (x),
the resulting linear control system can then be applied.
dx
and similarly, the Lie derivative of h(x) along g(x) as,

Feedback Linearization of SISO

Systems

Lg h(x) =

d h(x)
g(x).
dx

With this new notation, we may express y as,

Here, we consider the case of feedback linearization of

a single-input single-output (SISO) system. Similar results can be extended to multiple-input multiple-output
(MIMO) systems. In this case, u R and y R . We
wish to nd a coordinate transformation z = T (x) that
transforms our system (1) into the so-called normal form
which will reveal a feedback law of the form

y = Lf h(x) + Lg h(x)u
Note that the notation of Lie derivatives is convenient
when we take multiple derivatives with respect to either
the same vector eld, or a dierent one. For example,
1

L2f h(x)

d(Lf h(x))
= Lf Lf h(x) =
f (x),
dx

and

FEEDBACK LINEARIZATION OF SISO SYSTEMS

z1 (x)
z2 (x)

z = T (x) = . =
..

y
y
..
.

h(x)
Lf h(x)
..
.

n1
(n1)
Lf h(x)
y

zn (x)
d(Lf h(x))
g(x).
dx

transforms trajectories from the original x coordinate

system into the new z coordinate system. So long as this
transformation is a dieomorphism, smooth trajectories
in
the original coordinate system will have unique coun1.2 Relative degree
terparts in the z coordinate system that are also smooth.
In our feedback linearized system made up of a state vec- Those z trajectories will be described by the new system,
tor of the output y and its rst (n1) derivatives, we must
understand how the input u enters the system. To do this,
z1 = Lf h(x) = z2 (x)

we introduce the notion of relative degree. Our system

z
2 = L2f h(x) = z3 (x)
given by (1) and (2) is said to have relative degree r W
.
..
at a point x0 if,

zn = Lnf h(x) + Lg Ln1

h(x)u
k
f
Lg Lf h(x) = 0
x in a neighbourhood of
x0 and all k r 2
Hence, the feedback control law
r1
Lg Lf h(x0 ) = 0
Lg Lf h(x) =

Considering this denition of relative degree in light of

the expression of the time derivative of the output y , we
can consider the relative degree of our system (1) and (2)
to be the number of times we have to dierentiate the
output y before the input u appears explicitly. In an LTI
system, the relative degree is the dierence between the
degree of the transfer functions denominator polynomial
(i.e., number of poles) and the degree of its numerator
polynomial (i.e., number of zeros).

1.3

Linearization by feedback

u=

1
(Lnf h(x)
n1
Lg Lf h(x)

+ v)

renders a linear inputoutput map from v to z1 = y . The

resulting linearized system

z1

z2

zn

= z2
= z3
..
.
=v

is a cascade of n integrators, and an outer-loop control v

For the discussion that follows, we will assume that the may be chosen using standard linear system methodology.
relative degree of the system is n . In this case, after In particular, a state-feedback control law of
dierentiating the output n times we have,
y = h(x)
y = Lf h(x)
y = L2f h(x)
..
.
y (n1) = Ln1
h(x)
f
y (n) = Lnf h(x) + Lg Ln1
h(x)u
f

v = Kz

where the state vector z is the output y and its rst (n1)
derivatives, results in the LTI system

z = Az
with,

where the notation y (n) indicates the n th derivative of y

. Because we assumed the relative degree of the system

is n , the Lie derivatives of the form Lg Lif h(x) for i =

A=
1, . . . , n2 are all zero. That is, the input u has no direct

contribution to any of the rst (n 1) th derivatives.

0
0
..
.

1
0
..
.

0
1
..
.

0
k1

0
k2

0
k3

...
...
..
.

0
0
..
.

...
1
. . . kn

The coordinate transformation T (x) that puts the system

into normal form comes from the rst (n1) derivatives. So, with the appropriate choice of k , we can arbitrarily
In particular,
place the closed-loop poles of the linearized system.

1.4

Unstable zero dynamics

Feedback linearization can be accomplished with systems

that have relative degree less than n . However, the normal form of the system will include zero dynamics (i.e.,
states that are not observable from the output of the system) that may be unstable. In practice, unstable dynamics
may have deleterious eects on the system (e.g., it may
be dangerous for internal states of the system to grow unbounded). These unobservable states may be controllable
or at least stable, and so measures can be taken to ensure
these states do not cause problems in practice.

See also
Nonlinear control

Further reading
A. Isidori, Nonlinear Control Systems, third edition,
Springer Verlag, London, 1995.
H. K. Khalil, Nonlinear Systems, third edition, Prentice Hall, Upper Saddle River, New Jersey, 2002.
M. Vidyasagar, Nonlinear Systems Analysis second
edition, Prentice Hall, Englewood Clis, New Jersey, 1993.
B. Friedland, Advanced Control System Design Facsimile edition, Prentice Hall, Upper Saddle river,
New Jersey, 1996.

External links
ECE 758: Modeling and Nonlinear Control of a
Single-link Flexible Joint Manipulator Gives explanation and an application of feedback linearization.
ECE 758: Ball-in-Tube Linearization Example
Trivial application of linearization for a system already in normal form (i.e., no coordinate transformation necessary).
Wolfram language functions to do feedback linearization, compute relative orders, and determine
zero dynamics.

5 TEXT AND IMAGE SOURCES, CONTRIBUTORS, AND LICENSES

Text and image sources, contributors, and licenses

5.1

Text

Feedback linearization Source: https://en.wikipedia.org/wiki/Feedback_linearization?oldid=695239612 Contributors: Giftlite, TedPavlic, Cooperised, BD2412, Encyclops, Colonies Chris, Jmdaly, Bwpach, CmdrObot, Jiuguang Wang, Pleasantville, Addbot, Zehnn,
ZroBot, Pivs, Chrisdembia, Applebytom and Anonymous: 8

5.2

Images

File:Feedback_linearization.svg Source: https://upload.wikimedia.org/wikipedia/commons/9/97/Feedback_linearization.svg License:

CC BY-SA 3.0 Contributors: Own work Original artist: Chris Dembia and Tom Uchida
File:Text_document_with_red_question_mark.svg Source: https://upload.wikimedia.org/wikipedia/commons/a/a4/Text_document_
with_red_question_mark.svg License: Public domain Contributors: Created by bdesham with Inkscape; based upon Text-x-generic.svg
from the Tango project. Original artist: Benjamin D. Esham (bdesham)

5.3

Content license

Creative Commons Attribution-Share Alike 3.0