LMIs in Control/pages/MatrixEigenValueMinimization
LMI for Minimizing Eigenvalue of a Matrix
Synthesizing the eigenvalues of a matrix plays an important role in designing controllers for linear systems. The eigenvalues of the state matrix of a linear time-invariant system determine if the system is stable or not. The system is stable if all the eigenvalues of the state matrix are located in the left half of the complex plane. Thus, we may desire to minimize the maximal eigenvalue of the state matrix such that the minimized eigenvalue is placed in the left half-plane, which guarantees that the system is stable.
The System[edit | edit source]
Assume that we have a matrix function of variables :
where are symmetric matrices.
The Data[edit | edit source]
The symmetric matrices () are given.
The Optimization Problem[edit | edit source]
The optimization problem is to find the variables to minimize the following cost function:
where is the cost function and indicates the maximim eigenvalue of a matrix.
According to Lemma 1.1 in LMI in Control Systems Analysis, Design and Applications (page 10), the following statements are equivalent
where is defined as the maximim eigenvalue of the matrix .
The LMI: LMI for eigenvalue minimization[edit | edit source]
This optimization problem can be converted to an LMI problem.
The mathematical description of the LMI formulation can be written as follows:
Conclusion:[edit | edit source]
As a result, the variables after solving this LMI problem.
Moreover, we obtain the maximum eigenvalue, , of the matrix .
Implementation[edit | edit source]
A link to Matlab codes for this problem in the Github repository:
Related LMIs[edit | edit source]
External Links[edit | edit source]
-  - LMI in Control Systems Analysis, Design and Applications
- Eigenvalues and Eigenvectors of a Matrix