Introduction to Linear and Nonlinear Observers

Zoran Gajic, Rutgers University

Part 1 Review Basic Observability (Controllability) Results

Part 2 Introduction to Full- and Reduced-Order Linear Observers

Part 3 Introduction to Full- and Reduced-Order Nonlinear Observers

PART 1: BASIC OBSERVABILITY (CONTROLLABILITY) RESULTS Observability Theorem in Discrete-Time The linear discrete-time system with the corresponding measurements

is observable if and only if the observability matrix

. . .

has rank equal to

.
2

 

Observability Theorem in Continuous-Time

The linear continuous-time system with the corresponding measurements

is observable if and only if the observability matrix

has full rank equal to

  

. . .

 %$ " &# ! 
3

Controllability Theorem in Discrete-Time

The linear discrete-time system

is controllable if and only if the controllability matrix

has full rank equal to

4 1 5( 2( '

) '(

'

'

. . .

. . .

. . .

'
dened
4

' ' ' '

Controllability Theorem in Continuous-Time

The linear continuous-time system

is controllable if and only if the controllability matrix

dened by

has full rank equal to

C @ 57 A7

9 7 8

. . .

. . .

. . .

D
5

Similarity Transformation For a given system

we can introduce a new state vector

by a linear coordinate transformation as

where

is some nonsingular

matrix. A new state space model is obtained as

where

G F

E G F G HF
6

Eigenvalue Invariance Under a Similarity Transformation

A new state space model obtained by the similarity transformation does not change internal structure of the model, that is, the eigenvalues of the system remain the same. This can be shown as follows

Note that in this proof the following properties of the matrix determinant have been used

P I P
7

P I

P I R P I Q P

Controllability Invariance Under a Similarity Transformation The pair is controllable if and only if the pair is controllable.

This theorem can be proved as follows

Since we get

is a nonsingular matrix (it cannot change the rank of the product

U S T

. . .

. . .

. . .

U T
),
8

U S T

U T

. . .

U S T
. . .

. . .

. . .

. . .

. . .

Observability Invariance Under a Similarity Transformation The pair is observable if and only if the pair is observable.

The proof of this theorem is as follows

that is,

The nonsingularity of

implies

Y W X

Y X

Y X

Y W X

Y X

Y W X

. . .

. . .

. . .

Y X V
9

Y X V Y X

Y X

Y X Y X V

PART 2: INTRODUCTION TO LINEAR OBSERVERS

Sometimes all state space variables are not available for measurements, or it is not practical to measure all of them, or it is too expensive to measure all state space variables. In order to be able to apply the state feedback control to a system, all of its state space variables must be available at all times. Also, in some control system applications, one is interested in having information about system state space variables at any time instant. Thus, one is faced with the problem of estimating system state space variables. This can be done by constructing another dynamical system called the observer or estimator, connected to the system under consideration, whose role is to produce good estimates of the state space variables of the original system.

10

The theory of observers started with the work of Luenberger (1964, 1966, 1971) so that observers are very often called Luenberger observers. According to Luenberger, any system driven by the output of the given system can serve as an observer for that system. Two main techniques are available for observer design. The rst one is used for the full-order observer design and produces an observer that has the same dimension as the original system. The second technique exploits the knowledge of some state space variables available through the output algebraic equation (system measurements) so that a reduced-order observer is constructed only for estimating state space variables that are not directly obtainable from the system measurements.

11

Full-Order Observer Design Consider a linear time invariant continuous system

appropriate dimensions. Since the system output variables,

at all times, we may construct another articial dynamic system of order for example, of capacitors and resistors) having the same matrices

and compare the outputs

and

.
12

where

with constant matrices

a
having , are available (built,

These two outputs will be different since in the rst case the system initial condition is unknown, and in the second case it has been chosen arbitrarily. The difference between these two outputs will generate an error signal

which can be used as the feedback signal to the articial system such that the estimation (observation) error is reduced as much as possible,

hopefully to zero (at least at steady state). This can be physically realized by proposing the system-observer structure as given in the next gure.

13

System

Ce

x x

System-observer structure

In this structure

represents the observer gain and has to be chosen such that

the observation error is minimized. The observer alone is given by

Observer

y=Cx

e f

y=Cx

+ -

14

Remark 1: Note that the observer has the same structure as the system plus the driving feedback term that contain information about the observation error

The role of the feedback term is to reduce the observation error to zero (at steady state). Remark 2: The observer is usually implemented on line as a dynamic system driven by the same input as the original system and the measurements coming from the original systems, that is (note )

15

It is easy to derive an expression for dynamics of the observation error as

If the observer gain

is chosen such that the feedback matrix

is

asymptotically stable, then the estimation error

will decay to zero for any is observable.

More precisely, by taking the transpose of the estimation error feedback matrix, i.e. , we see that if the pair

its poles in arbitrarily asymptotically stable positions. Note that controllability of the pair

is equal to observability of the pair

the observability and controllability matrices.

r s

initial condition

. This can be achieved if the pair

s s

is controllable, then we can locate

, see expressions for

16

In practice the observer poles should be chosen to be about ten times faster than the system poles. This can be achieved by setting the minimal real part of observer eigenvalues to be ten times bigger than the maximal real part of system eigenvalues, that is

(in practice 10 can be replace by 5 or 6). Theoretically, the observer can be made arbitrarily fast by pushing its eigenvalues far to the left in the complex plane, but very fast observers generate noise in the system.

t #
17

&y

wu xvt

System-Observer Conguration We will show that the system-observer structure preserves the closed-loop system poles that would have been obtained if the linear perfect state feedback control had been used. The system under the perfect state feedback control, that is has the closed-loop form as

so that the eigenvalues of the matrix under perfect state feedback.

are the closed-loop system poles

In the case of the system-observer structure, as given in the given block diagram, we see that the actual control applied to both the system and the observer is given by

18

By eliminating

and

from the

augmented system-observer conguration, we obtain the following closed-loop form

What are the eigenvalues of this augmented system? If we write the system-error equation, we have

Since the state matrix of this system is upper block triangular, its eigenvalues are equal to the eigenvalues of matrices exists among and and . A very simple relation

19

Note that the matrix

is nonsingular. In order to go from

-coordinates to

coordinates we have to use the similarity transformation, which preserves the same eigenvalues, that is -coordinates. Separation Principle This important observation that the system-observer conguration has closedloop poles separated into the original system closed-loop poles obtained under perfect state feedback, , and the actual observer closed-loop poles, and , are also the eigenvalues in the

, is known as the separation principle. Hence, we can independently design the system poles using the system feedback gain gain and independently design the observer poles using the observer feedback .

20

t Clock To Workspace2 x = Ax+Bu y = Cx+Du Linear System (state space form) y(t)

y system output y(t)

Mux

F* u feedback gain F Observer (state space form)

System-Observer Conguration in SIMULINK

x = Ax+Bu y = Cx+Du

xhat(t) C* u matrix C yhat observer output yhat(t)

21

should be set (by clicking on and opening the observer state space block) as >> A=A; B=B; C=C; D=zeros(p,r); % assuming D=0 >> % to be able to run simulation you must assign any value to the system initial >> % condition since in practice this value is given, but unknown, that is >> x0 = any vector of dimension n Since the observer is implemented as

the observer state space matrices in SIMULINK should be specied (by clicking on and opening the observer state space block) as >> Aobs=A-K*C; Bobs=[B K]; Cobs=eye(n); Dobs=zeros(n,r+p); >> xobs=any n-dimensional vector
22

Since

, the state space form for the system matrices

Discrete-Time Full-Order Observer The same procedure can be applied to in the discrete-time domain producing the analogous results. Discrete-time system:

Discrete-time observer:

Observation error dynamics (

):

h e

e e f e e e e
23

is chosen to make the observer stable,

faster than the system, which requires

In practice, the observer should be six to ten times faster than the system. Closed-loop system-observer conguration

The system-error dynamic

The separation principle holds also.

i i i i i i i i

i i i i i i i

, and much

Reduced-Order Observer (Estimator) Consider the linear system with the corresponding measurements

We will show how to derive an observer of reduced dimensions by exploiting knowledge of the output measurement equation. Assume that the output matrix has rank , which means that the output equation represents algebraic equations. Thus, equation linearly independent

produces

algebraic equations for

unknowns of

observer of order

for estimation of the remaining

k
. Our goal is to construct an state space variables.
25

In order to simplify derivations and without loss of generality, we will consider the linear system with the corresponding measurements dened by

This is possible since it is known from linear algebra that if

Hence, mapping the system in the new coordinates via the similarity transformation, we obtain the given structure for the measurement matrix.

r q

then it exists a nonsingular matrix

such that

, which implies

p o n

m
26

l n r n p o 2p r Hq r s r

Partitioning compatibly the system equation, we have

The state variables

are directly measured (observed) at all times, so that

. To construct an observer for

, we use the knowledge that

an observer has the same structure as the system plus the driving feedback term whose role is to reduce the estimation error to zero. Hence, an observer for

Since

does not carry information about

, this observer will not be able

to reduce the corresponding observation error to zero,

t u
is .
27

t u

t u u u t vu t t u t

uu u t

tvu t t

t u

However, if we differentiate the output variable we get

that is

carries information about

. The reduced-order observer with the is

feedback information coming from

The observation error dynamics can be obtained from

To place the reduced-observer poles arbitrarily (the reduced-order observer must be

x w {

x x{

stable and much faster than the system), we need

controllable.
28

w x w x
as

x yw x x w x yw x x x x

x zx

w w w x x x w w w x

w w vx w x

By duality between controllability and observability, is dual to observability of

controllability of

It is easy to show using the Popov-Belevitch observability test

that

observable implies

Hence, if the original system is observable, we can construct the reduced-order observer whose observation error will decay quickly to zero.

} ~ 

} } } ~ 

} }

} ~ |

} }|

29

Proof of the claim

observable implies

30

The need for

in the reduced-order observer equation

can be eliminated by introducing the change of variables leads to

, which

31

Reduced-Order Observer Derivation without a Change of Coordinates Consider the linear system with the corresponding measurements

Assume that the output matrix represents

has rank , which means that the output equation

linearly independent algebraic equations. Thus, equation

produces

algebraic equations for

unknowns of

observer of order

for estimation of the remaining

. Our goal is to construct an state space variables.

The procedure for obtaining this observer is not unique, which is obvious from

Now, we have

Since the vector

is unknown, we will construct an observer to estimate it.

and introduce a vector

as

33

the next step. Assume that a matrix

exists such that

Introduce the notation

so that

An observer for , that is

can be constructed by nding rst a differential equation for

Note that from this system we are not able to construct an observer for does not contain explicit information about the vector .

since
34

To see this, we rst observe that

The measurements

are given by

35

If we differentiate the output variable we get

i.e.

carries information about

. An observer for

the last two equations as

where

is the observer gain. If in the differential equation for

by its estimate, we will have

is obtained from

we replace

36

This produces the following observer for

Since it is impractical and undesirable to differentiate

in order to get

(this operation introduces noise in practice), we take the change of variables

This leads to an observer for

of the form

where

37

The estimates of the original system state space variables are now obtained as

The obtained system-reduced-observer structure is presented in the next gure.

Bq Reduced observer

L2

System-reduced-observer structure 38

q q

B Kq

System

L1+L2K1 + +

Setting Reduced-Order-Observer Eigenvalues in the Desired Location We need that the eigenvalues of the reduced-order observer

be roughly ten times faster than the closed-loop system eigenvalues determined

the rst

state variables are directly measured). This is dual to the requirement is controllable.

proved similarly to the proof of the claim

observable implies

Note that it can be shown that

observable implies

(analogous result to the requirement

observable for the case when

by

. This can be done if the pair

is observable and .
39

We can set the reduced-observer eigenvalues using the following MATLAB statements: >> % checking the observability condition >> O=obsv(C1*A*L2,C*A*L2); >> rank(O); % must be equal to p >> % nding the closed-loop system poles >> lamsys=eig(A-B*F); maglamsys=abs(real(lamsys)) >> % nding the closed-loop reduced-order observer poles >> % input desired lamobs (reduced-order observer eigenvalues) >> K1T=place((C1*A*L2),(C*A*L2),lamobs); >> K1=K1T

40

PART 3 INTRODUCTION TO NONLINEAR OBSERVERS

We have seen that to observe the state of the linear system dened by

we construct a linear observer that has the same structure as the system plus the driving feedback term whose role is to reduce the observation error to zero

Studying observers for nonlinear systems is theoretically much harder. However, we can use the same logic to construct a nonlinear observer.

41

Consider a nonlinear controlled system with measurements

of dimensions

and

Based on the knowledge of linear observers, we can propose the following structure for a nonlinear observer

Hence, the nonlinear observer is dened by

and

are nonlinear vector functions, respectively,

42

The observer gain and , that is,

is a nonlinear matrix function that in general depends on . It has to be chosen such that the observation error, tends to zero (at least at steady state).

The observation error dynamics is determined by

By eliminating

from the error equation, we obtain

At the steady state we have

43

It is obvious that

is the solution of this algebraic equation, which indiat steady state. The gain

cates that the constructed observer may have

must be chosen such that the observer and error dynamics are asymptotically stable (to force the error at steady state to ).

The asymptotic stability will be examined using the rst stability method of Lyapunov. The Jacobin matrix for the error equation is given by

By the rst stability method of Lyapunov, the Jacobian matrix must have all eigenvalues in the left half plane for all working conditions, that is for all and , where and are the sets of admissible state and control variables.

44

The error dynamics asymptotic stability condition is

Similarly, for the observer we have

and it is required that the observer is also asymptotically stable

45

55

55

Nonlinear observer block diagram is presented in the next gure

46

Reduced-Order Nonlinear Observers

variables

Let us partition compatible the state equations

The estimate for the state variables can be obtained as

construct a nonlinear observer to estimate the remaining

Assume that

state variables are directly measured and we need to state

47

Let us assume that the dynamic system (observer) for

has the following form

We have to nd the reduced-order observer gain structure dened by at steady state.

such that the observation error

The dynamic equation for the error is obtained as follows

Since our goal is that at steady state

, we have

and the reduced-order observer tends to zero

48

Hence, the reduced-order observer structure is given by

The error dynamic must be asymptotically stable

which means that by the rst method of Lyapunov the Jacobian matrix must have all eigenvalues in the left half plane for all working conditions, that is for all and , where

and

are the sets of admissible state and control variables.

49

The error dynamics asymptotic stability require that

Similarly, the reduced-order observer dynamics must be asymptotically stable. The block diagram of the reduced-order nonlinear observer is given below

50

z

This lecture on observers is prepared using the following literature: [1] Z. Gajic and M. Lelic, Modern Control Systems Engineering, Prentice Hall International, London, 1996, (pages 241247 on full- and reduced-order observers). [2] Stefani, Shahian, Savant and Hostetter, Design of Feedback Systems, Oxford University Press, New York, 2002, (pages 650652 on reduced order linear observer). [3] B. Friedland, Advanced Control System Design, Prentice Hall, Englewood Cliffs, 1996 (pages 164166 and 174175, 183187 on full- and reduced-order nonlinear observers). Basic results on observability (controllability) are reviewed from [1].

51