Difference between revisions of "Output Feedback"

From FBSwiki
Jump to: navigation, search
(→‎Chapter Summary)
(→‎Chapter Summary: updated equations to amsmath)
Line 30: Line 30:
 
<ol>
 
<ol>
 
<li> <p> A linear system with dynamics
 
<li> <p> A linear system with dynamics
−
<center>
+
<center><amsmath>
−
{|
+
\begin{aligned}
−
|-
+
  \dot x &= A x + B u &\quad x &\in {R}^n, u \in {R} \\
−
| <math> \dot x = A x + B u </math>
+
  y &= C x + D u & y &\in {R}
−
| &nbsp;&nbsp;
+
\end{aligned}
−
| <math> x \in R^n, u \in R </math>
+
</amsmath></center>
−
|-
+
−
| <math> y = \left. C x + D u \right.</math>
+
−
| &nbsp;&nbsp;
+
−
| <math> y \in R</math>
+
−
|}
+
−
</center>
+
 
is said to be ''observable'' if we can determine the state of the system through measurements of the input <math>u(t)</math> and the output <math>y(t)</math> over a time interval <math>[0, T]</math>.
 
is said to be ''observable'' if we can determine the state of the system through measurements of the input <math>u(t)</math> and the output <math>y(t)</math> over a time interval <math>[0, T]</math>.
 
</p>
 
</p>
  
 
<li><p>The ''observability matrix'' for a linear system is given by
 
<li><p>The ''observability matrix'' for a linear system is given by
−
<center><math>
+
<center><amsmath>
−
W_o = \left[\begin{matrix} C \\ CA \\ \vdots \\ C A^{n-1} \end{matrix}\right].
+
W_o = \begin{bmatrix} C \\ CA \\ \vdots \\ C A^{n-1} \end{bmatrix}.
−
</math></center>
+
</amsmath></center>
 
A linear system is observable if and only if the observability matrix <math>W_o</math> is full rank.  Systems that are not reachable have "hidden" states that cannot be determined by looking at the inputs and outputs.
 
A linear system is observable if and only if the observability matrix <math>W_o</math> is full rank.  Systems that are not reachable have "hidden" states that cannot be determined by looking at the inputs and outputs.
 
</p>
 
</p>
  
 
<li><p>A linear system of the form
 
<li><p>A linear system of the form
−
<center>
+
<center><amsmath>
−
{|
+
\begin{aligned}
−
|-
+
  \frac{dz}{dt} &= \begin{bmatrix}
−
| align=right |<math> \frac{dz}{dt} = </math>
+
−
| align = left | <math>
+
−
  \left[\begin{matrix}
+
 
       -a_1 & 1 & 0 & \cdots & 0 \\
 
       -a_1 & 1 & 0 & \cdots & 0 \\
 
       -a_2 & 0 & 1 &      & 0\\  
 
       -a_2 & 0 & 1 &      & 0\\  
Line 64: Line 55:
 
       -a_{n-1} & 0 & 0 &  & 1 \\
 
       -a_{n-1} & 0 & 0 &  & 1 \\
 
       -a_n & 0 & 0 &  & 0\\
 
       -a_n & 0 & 0 &  & 0\\
−
   \end{matrix}\right] z+ \left[\begin{matrix}
+
   \end{bmatrix} z+ \begin{bmatrix}
 
       b_1 \\  b_2 \\  \vdots \\ b_{n-1}  \\  b_n
 
       b_1 \\  b_2 \\  \vdots \\ b_{n-1}  \\  b_n
−
   \end{matrix}\right] u
+
   \end{bmatrix} u \\
−
</math>
+
  y &= \begin{bmatrix}
−
|-
+
−
| align=right |  <math>\left. y = \right. </math>
+
−
| <math>\left[\begin{matrix}
+
 
     1 & 0 & 0 \cdots & 0
 
     1 & 0 & 0 \cdots & 0
−
     \end{matrix}\right] z+ Du.
+
     \end{bmatrix} z+ Du.
−
</math>
+
\end{aligned}
−
|}
+
</amsmath></center>
−
</center>
+
 
is said to be in ''observable canonical form''.  A system in this form is always observable and has a characteristic polynomial given by
 
is said to be in ''observable canonical form''.  A system in this form is always observable and has a characteristic polynomial given by
−
<center><math>
+
<center><amsmath>
 
   \det(sI-A) = s^n+a_1 s^{n-1} + \cdots + a_{n-1}s + a_n,
 
   \det(sI-A) = s^n+a_1 s^{n-1} + \cdots + a_{n-1}s + a_n,
−
</math></center>
+
</amsmath></center>
 
An observable linear system can be transformed into observable canonical form through the use of a coordinate transformation <math>z = T x</math>.   
 
An observable linear system can be transformed into observable canonical form through the use of a coordinate transformation <math>z = T x</math>.   
 
</p>
 
</p>
  
 
<li><p>An ''observer'' is a dynamical system that estimates the state of another system through measurement of inputs and outputs.  For a linear system, the observer given by
 
<li><p>An ''observer'' is a dynamical system that estimates the state of another system through measurement of inputs and outputs.  For a linear system, the observer given by
−
<center><math>
+
<center><amsmath>
 
\frac{d\hat x}{dt} = A \hat x + B u + L(y - C \hat x)
 
\frac{d\hat x}{dt} = A \hat x + B u + L(y - C \hat x)
−
</math></center>
+
</amsmath></center>
 
generates an estimate of the state that converges to the actual state if <math>A - LC</math> is has eigenvalues with negative real part.
 
generates an estimate of the state that converges to the actual state if <math>A - LC</math> is has eigenvalues with negative real part.
 
If a system is observable, then there exists a an ''observer gain'' <math>L</math> such that the observer error is governed by a linear differential equation with an arbitrary characteristic polynomial.  Hence the eigenvalues of the error dynamics for an observable linear system can be placed arbitrarily through the use of an appropriate observer gain.
 
If a system is observable, then there exists a an ''observer gain'' <math>L</math> such that the observer error is governed by a linear differential equation with an arbitrary characteristic polynomial.  Hence the eigenvalues of the error dynamics for an observable linear system can be placed arbitrarily through the use of an appropriate observer gain.
Line 92: Line 79:
  
 
<li><p>A state feedback controller and linear observer can be combined to form a stabilizing controller for a reachable and observable linear system by using the estimate of the state in the feedback control law.  The resulting controller is given by
 
<li><p>A state feedback controller and linear observer can be combined to form a stabilizing controller for a reachable and observable linear system by using the estimate of the state in the feedback control law.  The resulting controller is given by
−
<center>
+
<center><amsmath>
−
{|
+
\begin{aligned}
−
|-
+
  \frac{d\hat x}{dt} &= A \hat x + B u + L(y - C \hat x) \\
−
| align=right | <math>\frac{d\hat x}{dt} = </math>
+
  u &= -K \hat x + K_r r
−
| align=left | <math>A \hat x + B u + L(y - C \hat x)</math>
+
\end{aligned}
−
|-
+
</amsmath></center>
−
| align = right | <math>\left. u = \right. </math>
+
−
| align = left | <math> \left. -K \hat x + K_r r \right. </math>
+
−
|}
+
−
</center>
+
 
</p>
 
</p>
  
 
<li><p>A discrete time, linear process with noise is given by
 
<li><p>A discrete time, linear process with noise is given by
−
<center>
+
<center><amsmath>
−
{|
+
\begin{aligned}
−
|-
+
  x(k+1) &= A x(k) + B u(k) + v(k) &\quad x &\in R^n, u \in R \\
−
| <math> \left. x(k+1) = A x(k) + B u(k) + v(k) \right. </math>
+
  y(k) &= C x(k) + D u(k) + w(k) & y &\in R
−
| &nbsp;&nbsp;
+
\end{aligned}
−
| <math> x \in R^n, u \in R </math>
+
</amsmath></center>
−
|-
+
−
| <math> y(k) = \left. C x(k) + D u(k) + w(k) \right.</math>
+
−
| &nbsp;&nbsp;
+
−
| <math> y \in R</math>
+
−
|}
+
−
</center>
+
 
where <math>v</math> is a vector, white, Gaussian random process with mean 0, autocovariance <math>R_w</math>, <math>w</math> is a white, Guassian random process with mean 0, variance <math>R_v</math>.  We take the initial condition to be random with mean 0 and covariance <math>P_0</math>. The optimal estimator is given by
 
where <math>v</math> is a vector, white, Gaussian random process with mean 0, autocovariance <math>R_w</math>, <math>w</math> is a white, Guassian random process with mean 0, variance <math>R_v</math>.  We take the initial condition to be random with mean 0 and covariance <math>P_0</math>. The optimal estimator is given by
−
<center><math>
+
<center><amsmath>
 
\hat x(k+1) = A \hat x(k) + B u(k) + L(y(k) - C \hat x(k))
 
\hat x(k+1) = A \hat x(k) + B u(k) + L(y(k) - C \hat x(k))
−
</math></center>
+
</amsmath></center>
 
where the observer gain satisfies
 
where the observer gain satisfies
 
<center><amsmath>
 
<center><amsmath>

Revision as of 15:13, 2 January 2007

Prev: State Feedback Chapter 7 - Output Feedback Next: Transfer Functions

In the last chapter we considered the use of state feedback to modify the dynamics of a system through feedback. In many applications, it is not practical to measure all of the states directly and we can measure only a small number of outputs (corresponding to the sensors that are available). In this chapter we show how to use output feedback to modify the dynamics of the system, through the use of state estimators (also called "observers"). We introduce the concept of observability and show that if a system is observable, itis possible to recover the state from measurements of the inputs andoutputs to the system.

Textbook Contents

Output Feedback (pdf, 16Sep06)

  • 1. Observability
  • 2. Observers
  • 3. Output Feedback
  • 4. A Design Example (missing)
  • 5. Additional Topics
  • 6. Further Reading
  • 7. Exercises

Lecture Materials

Supplemental Information

Chapter Summary

This chapter describes how to estimate the state of a system through measurements of its inputs and outputs:

  1. A linear system with dynamics

    math

    is said to be observable if we can determine the state of the system through measurements of the input and the output over a time interval .

  2. The observability matrix for a linear system is given by

    math

    A linear system is observable if and only if the observability matrix is full rank. Systems that are not reachable have "hidden" states that cannot be determined by looking at the inputs and outputs.

  3. A linear system of the form

    math

    is said to be in observable canonical form. A system in this form is always observable and has a characteristic polynomial given by

    math

    An observable linear system can be transformed into observable canonical form through the use of a coordinate transformation .

  4. An observer is a dynamical system that estimates the state of another system through measurement of inputs and outputs. For a linear system, the observer given by

    math

    generates an estimate of the state that converges to the actual state if is has eigenvalues with negative real part. If a system is observable, then there exists a an observer gain such that the observer error is governed by a linear differential equation with an arbitrary characteristic polynomial. Hence the eigenvalues of the error dynamics for an observable linear system can be placed arbitrarily through the use of an appropriate observer gain.

  5. A state feedback controller and linear observer can be combined to form a stabilizing controller for a reachable and observable linear system by using the estimate of the state in the feedback control law. The resulting controller is given by

    math

  6. A discrete time, linear process with noise is given by

    math

    where is a vector, white, Gaussian random process with mean 0, autocovariance , is a white, Guassian random process with mean 0, variance . We take the initial condition to be random with mean 0 and covariance . The optimal estimator is given by

    math

    where the observer gain satisfies

    math

    This estimator is an example of a Kalman filter.

Homework

Frequently Asked Questions

Additional Information

Retrieved from "http://www.cds.caltech.edu/~murray/amwiki/index.php?title=Output_Feedback&oldid=3069"