Adaptive neural prescribed performance control for switched pure-feedback non-linear systems with input quantization

Abstract

Purpose

This paper aims to investigate an adaptive prescribed performance control problem for switched pure-feedback non-linear systems with input quantization.

Design/methodology/approach

By using the semi-bounded continuous condition of non-affine functions, the controllability of the system can be guaranteed. Then, a constraint variable method is introduced to ensure that the tracking error satisfies the prescribed performance requirements. Meanwhile, to avoid the design difficulties caused by the input quantization, a non-linear decomposition method is adopted. Finally, the feasibility of the proposed control scheme is verified by a numerical simulation example.

Findings

Based on neural networks and prescribed performance control method, an adaptive neural control strategy for switched pure-feedback non-linear systems is proposed.

Originality/value

The complex deduction and non-differentiable problems of traditional prescribed performance control methods can be solved by using the proposed error transformation approach. Besides, to obtain more general results, the restrictive differentiability assumption on non-affine functions is removed.

Citation

Cao, Z., Zhang, L., Ahmad, A.M., Alsaadi, F.E. and Alassafi, M.O. (2022), "Adaptive neural prescribed performance control for switched pure-feedback non-linear systems with input quantization", Assembly Automation, Vol. 42 No. 6, pp. 869-880. https://doi.org/10.1108/AA-05-2022-0126