Robust Adaptive Trajectory Tracking Sliding Mode Control for Industrial Robot Manipulator using Fuzzy Neural Network

Abstract

This paper presents a control method for a two-link industrial robot manipulator system that uses Fuzzy Neural Networks (FNNs) based on Sliding Mode Control (SMC) to investigate joint position control for periodic motion and predefined trajectory tracking control. The proposed control scheme addresses the challenges of designing a suitable control system that can achieve the required approximation errors while ensuring the stability and robustness of the control system in the face of joint friction forces, parameter variations, and external disturbances. The control scheme uses four layers of FNNs to approximate nonlinear robot dynamics and remove chattering control efforts in the SMC system. The adaptive turning algorithms of network parameters are derived using a projection algorithm and the Lyapunov stability theorem. The proposed control scheme guarantees global stability and robustness of the control system, and position is proven. Simulation and experiment results from a two-link IRM in an electric power substation are presented in comparison to PID and AF control to demonstrate the superior tracking precision and robustness of the proposed intelligent control scheme.