Data-Driven Tuning of Stable Minimum Phase H∞ Controller

Abstract

Data-driven controller tuning methods directly design controllers using input/output data sets of a plant without using a mathematical model. This paper proposes a data-driven controller tuning method based on the H∞ control problem. The controller designed by the proposed method satisfies the H∞ criterion and stabilizes the closed-loop system. Furthermore, the proposed method derives a stable minimum phase controller. The controller is designed or tuned by iterative convex optimization. The effectiveness of the proposed design method is confirmed through comparison with the standard model-based H∞ control problem by simulations.