Neural network compensator-based robust iterative learning control scheme for mobile robots nonlinear systems with disturbances and uncertain parameters

Aiming at the problem of trajectory tracking control for mobile robot nonlinear systems with non-repetitive uncertain parameters, we propose a novel neural network compensator-based robust iterative learning control (NNRILC) scheme to achieve excellent tracking performance and uncertainty compensati...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Veröffentlicht in:Applied mathematics and computation 2024-05, Vol.469, p.128549, Article 128549
Hauptverfasser: Chen, Zhengquan, Hou, Yandong, Huang, Ruirui, Cheng, Qianshuai
Format: Artikel
Sprache:eng
Schlagworte:
Online-Zugang:Volltext
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
Beschreibung
Zusammenfassung:Aiming at the problem of trajectory tracking control for mobile robot nonlinear systems with non-repetitive uncertain parameters, we propose a novel neural network compensator-based robust iterative learning control (NNRILC) scheme to achieve excellent tracking performance and uncertainty compensation. The NNRILC scheme consists of a parallel structure that includes a robust iterative learning control (RILC) term and a neural network (NN) compensator. The RILC term is used to ensure robust control performance and promise closed-loop stability. The compensator based on the neural network is employed to handle the uncertainties resulted from the nonlinear dynamics as well as the suppressed disturbances in the mobile robot nonlinear systems. Additionally, the fourth-order Runge-Kutta algorithm is employed to solve the state differential equation of the mobile robot. Consequently, The H∞ robust technique is used to construct an iterative learning control update rule to reduce the impact of disturbances. Meanwhile, the RILC scheme is designed to optimize the neural network initial parameters and weights of the NN compensator at each trial. Then the convergence of the proposed NNRILC scheme is analyzed, and the results are incorporated into the control update for the next process trial. Finally, the effectiveness of the proposed method is verified by mobile robot simulation. •A novel ILC scheme combining neural network compensator is designed to solve trajectory tracking problem.•The Runge-Kutta algorithm is introduced to solve the state differential equation and the controller equation.•A neural network compensator is constructed to compensate the suppressed noises, disturbances, and uncertainties.•The H∞ robust technique is used to suppress the noises, disturbances in the ILC process.
ISSN:0096-3003
1873-5649
DOI:10.1016/j.amc.2024.128549