Prescribed performance based data-driven adaptive sliding mode control for discrete-time nonlinear systems

Abstract

This research explores the control problem of discrete-time nonlinear systems with prescribed performance utilizing the data-driven method. Firstly, the equivalent data model of the system is constructed employing the full-form dynamic linearization technique. Secondly, a novel error transformation function is proposed to transform the constrained tracking error into the unconstrained equivalent error. Subsequently, the new sliding function is established to achieve the original tracking error converges to the predetermined asymmetric domain. In contrast to existing methods, the proposed approach has a more straightforward controller structure. Additionally, the introduction of pade approximation simplifies the theoretical analysis. Finally, the superiority of the proposed scheme is validated through simulation results on the single-link manipulator system.