3
\$\begingroup\$

Consider the discrete-time system described by

$$ x(k+1) = \begin{pmatrix}0 & 1 \\ -2 & -3\end{pmatrix}x(k) + \begin{pmatrix}0\\1\end{pmatrix}u(k) $$

a) Find a feedback state vector that assigns the closed-loop poles to \$0.5\$ and \$1/4\$

b) Design an observer so that the error between the state and its estimate goes to zero faster than 0.5.

c) Write down the equations of an observer-based control system.

Attempt:

For the first part of the question, I have set up the feadback matrix (A - BK) and computed the state feed back vector K = \$[-11/6 -23/6]\$.

For the second part I know the question is looking for the observer L, which is obtained by setting up the matrix (A - LC), but I dont know how to make sure it goes faster than 0.5.How would I do this?

For the last part I think I just need to multiply the determinants of the matrices (A - BK) and (A - LC) (so I need both K and L ) because K and L are supposed to be able to be designed independently, is this right?

Any Help Appreciated.

Note: Exam Revision, not Homework.

\$\endgroup\$

1 Answer 1

1
\$\begingroup\$

Your feedback matrix K seems to be off; it yields poles at {1/3, 1/2}

For single-input observable systems, you can apply pole placement to place the observer poles wherever you like (as with the state feedback poles).

You might want to review just what a state observer is so you can better understand what you're building. The equations of an observer-based control system will include the both the system state equations and the observer state equations.

\$\endgroup\$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.