Data-Driven H∞Control for Unknown Linear Time-Invariant Systems with Bounded Disturbances

Abstract

<p>Motivated by learning controllers directly from data, this paper studies the problem of designing a data-driven controller for unknown linear time-invariant (LTI) systems with bounded disturbances. An H-infinity controller is designed using a group of input/state/output data without any model information. The obtained results are further applied to the ARX systems using only input/output data. The main advantage of the proposed controller is that it can balance the trade-off between the robustness and control performance of the closed-loop system by choosing the number of data sequences and the length of each sequence. The effectiveness of the proposed method is shown using simulation results of an outdoor antenna servo system.<br></p>