Adaptive Compensation for Bilateral Teleoperation Systems With Asymmetric Time-Varying Delays Using Neural Network in P+d Control Framework

Abstract

Bilateral teleoperation systems face significant challenges in achieving accurate master-slave position convergence, especially under the influence of time-varying communication delays. This document presents an adaptive neural network compensation strategy within a proportional-derivative-like (P+d) control framework. The proposed controller addresses uncertainties associated with time delays and system dynamics using online adaptation of neural networks. Furthermore, a rigorous theoretical analysis of the closed-loop teleoperation system is presented and validated through numerical simulations on two-degree-of-freedom manipulators. Finally, the results demonstrate that adaptive neural networks can significantly improve the operational reliability of bilateral teleoperation systems in dynamic environments.