Gain Scheduled Output Feedback Controller with Low Computational Burden

Abstract

This paper proposes a synthesis method of output feedback gain scheduled controller, which minimizes the robust H2 norm of the closed loop system, with low computational burden in real time control. LMI framework allows synthesis method of gain scheduled controllers, by using change-of-variables, convex hull, and parameter dependent Lyapunov matrix. However, the synthesized controllers by the framework usually need to calculate inverse of the parameter-dependent Lyapunov matrix at each sampling at real time control. The calculation of inverse of matrix causes large computational burden. To reduce the computational burden, we propose an estimator based state feedback controller synthesis framework. A special structure for the feedback gain and the estimator gain is introduced to reduce the dimension of the parameter-dependent matrix, whose inverse is required in real time control. We use the redundant descriptor representation to handle the special structure in the synthesis condition in terms of LMIs. The effectiveness is verified by a numerical example.