TRAJECTORY TRACKING CONTROL FOR FLEXIBLE-JOINT MANIPULATOR WITH TIME-VARYING UNCERTAINTIES USING BACKSTEPPING AND CHEBYSHEV NEURAL NETWORK

The tracking control for flexible-joint manipulator system with time-varying uncertainties is investigated in this paper. The control performance of the system is inevitably affected by the mismatched uncertainties. To tackle this issue, a novel controller that integrates backstepping and Chebyshev neural networks (CNN) is proposed. Backstepping is used to deal with the mismatched problem, and CNN are used to approximate the nonlinear functions. The adaptive law can be derived from Lyapunov stability analysis and all the signals in closed-loop system are bounded. The comparative simulation experiments validate the superior performance of the proposed method over the commonly used RBF NN. © 2024 Acta Press. All rights reserved.