Title: A new data-driven sliding mode learning control for discrete-time MIMO linear systems

Authors: Lei Cao; Shouli Gao; Dongya Zhao

Addresses: College of New Energy, China University of Petroleum, Qingdao 266580, China ' College of New Energy, China University of Petroleum, Qingdao 266580, China ' College of New Energy, China University of Petroleum, Qingdao 266580, China

Abstract: A new data-driven sliding mode learning control (DDSMLC) is designed for a class of discrete-time MIMO linear systems in the presence of uncertainties. In this scheme, a new control is designed to enforce the states to reach and remain on the sliding surface. In addition, a recursive algorithm using system measured data is adopted to estimate the unknown system parameters, so a complete data-driven sliding mode control is designed, which does not need to know any parameters in the system. Moreover, the chattering is reduced because there is no non-smooth control used in DDSMLC. After the strict stability analysis, the effectiveness of DDSMLC is validated by MATLAB simulations.

Keywords: data-driven; discrete-time MIMO linear systems; parameter estimation algorithm; sliding mode learning control.

DOI: 10.1504/IJISE.2022.126074

International Journal of Industrial and Systems Engineering, 2022 Vol.42 No.2, pp.211 - 229

Received: 01 Apr 2020
Accepted: 15 Jan 2021
Published online: 10 Oct 2022 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article