Data Driven Robust Superstable Control of Switched Systems

摘要

This paper considers the problem of data driven control of switched linear MIMO systems. Our goal is, given experimental data collected at different operating points and apriori model structure, to directly design a controller that super-stabilizes all plants in the consistency set, for any arbitrary switching sequence, without explicitly identifying the plant. Our main result shows that this problem can be recast into a convex optimization problem and efficiently solved, leading to a robust controller with guaranteed l_∞ worse-case performance for any switching amongst all plants that could have generated the observed experimental data.