Data-driven Control of Constrained Systems

Abstract

In this paper, we propose a data-driven design of feedback/feedforward controllers to achieve a desired tracking performance for discrete-time constrained linear systems. The proposed methods can handle any time-domain linear constraints on the input, output, and (multiple) internal state variables. Based on the data-driven prediction of such signals, the design problems are reduced to quadratic optimization problems. The effectiveness of the methods is investigated through two numerical simulations of designing a feedback and feedforward controller, respectively.