Sliding Mode Control of Robot Manipulators Based on Neural Network Reaching Law

Abstract

A new neural network sliding mode control method of robot manipulators is proposed, which is formed by incorporating sliding mode variable structure control (SMVSC) and neural network reaching law. The reaching law parameters are regulated adaptively by two feedforward neural networks (FNNs) respectively. This method converts a multi-input system into <i xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">n</i> single-input systems. Its control arithmetic is simple and easy to implement. It can not only eliminate the chattering of sliding mode control and strengthen the system robustness, but also improve the character of reaching phase. Tracking errors can promptly converge to a neighborhood of zero. The simulation results of two-degree-of-freedom robot manipulators prove the effectiveness of this scheme.