Verifying stabilizing controllers for performance improvement using closed-loop data

Abstract

SUMMARY In iterative identification and control, one needs to determine the closed-loop performance of a new controller designed with the current model so that a decision can be made on whether a new iteration of plant identification and controller design is necessary. In multiple model adaptive control, at times, a decision to connect a different controller is made. It is desirable to check the closed-loop performance with a new controller before it is switched in. Consider a feedback interconnection of an unknown linear plant and a known linear stabilizing controller, and assume that some knowledge of the closed-loop system is available. We aim to develop theory and tools for experimentally verifying the closed-loop performance with a prospective stabilizing controller in advance. The simulation results demonstrate efficacy and versatility of our results.