Scribd
Upload
27 views

Quadratic Optimal Regulator Systems (LQR), (p793) : Minimize The Performance Index, Defined As

The document discusses the quadratic optimal regulator or LQR method for controlling systems. LQR offers more direct control over system performance compared to pole assignment methods. It involves minimizing a performance index that balances the state and control input over time, subject to the system dynamics. The optimal state feedback gain K is derived, resulting in an asymptotically stable closed-loop system where the optimal cost J depends only on initial conditions. MATLAB's lqr function can solve the LQR problem numerically since an analytical solution by hand is impractical. Examples from the text and problems are provided for additional practice.

Uploaded by

Abdullah Abu Lehyah
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as DOCX, PDF, TXT or read online on Scribd
27 views

Quadratic Optimal Regulator Systems (LQR), (p793) : Minimize The Performance Index, Defined As

The document discusses the quadratic optimal regulator or LQR method for controlling systems. LQR offers more direct control over system performance compared to pole assignment methods. It involves minimizing a performance index that balances the state and control input over time, subject to the system dynamics. The optimal state feedback gain K is derived, resulting in an asymptotically stable closed-loop system where the optimal cost J depends only on initial conditions. MATLAB's lqr function can solve the LQR problem numerically since an analytical solution by hand is impractical. Examples from the text and problems are provided for additional practice.

Uploaded by

Abdullah Abu Lehyah
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as DOCX, PDF, TXT or read online on Scribd
You are on page 1/ 1

Quadratic Optimal Regulator Systems (LQR), (p793)

The advantage of LQR methods over pole assignment methods is that LQR
offers a more direct control on system performance.
The LQR problem for real matrices is posed as follows:
Minimize the performance index J , defined as
∞
J=∫ ( xT Qx+ uT Ru ) dt
0

usingu=−Kx subject to the constraint ẋ= Ax+ Bu .


where Q is a positive definite or positive semi-definite n × n matrix,
R must be a positive definite m× m matrix, and the system is controllable.
The matrices Q, and R determine the relative importance of the error x , and the
expenditure of energy. The control vector u is assumed to be unconstrained.
Following the necessary assumptions (see book pp794-795) , it can be
shown that when u=−Kx , the optimal state feedback matrix K is
K opt =R−1 BT P

Where P is a unique positive definite matrix satisfying


AT P+ P A−P B R−1 BT P+Q=0

Resulting in J=x T ( 0 ) P x (0)

i.e. the value of J is solely dependent on the initial conditions since P is unique.
Furthermore, the optimal controller guarantees a closed loop system to be
asymptotically stable.
Solution by hand is not practical. Use the lqr function of MATLAB.
For examples: Study all examples in the book from pp796-802. Pay particular
attention to the proper use of the lqr function of MATLAB.
For additional Examples: A Solved problem, A-10-16, pp846-848.
For additional Problems: Unsolved problems B-10-19 to B-10-21, p 858.

You might also like