Optimal pole-placement for linear multi-input controllable systems

Abstract

This paper presents a new method for the design of linear state-feedback controllers for multiinput systems which results in a desired preselected set of distinct closed-loop eigenvalues and simultaneously optimizes a quadratic performance index. The method is based on a computationally efficient algorithm, which constructs the feedback gainmatrix by selecting the optimal closed-loop eigenvectors of the system without solving the algebraic matrix Riccati equation.