ANN-inversion based fractional-order sliding control for the industrial robot

Abstract

To improve the control performance of the industrial robot, an ANN-inversion based fractional-order sliding mode control(FOSMC) scheme is proposed. Firstly, the BP neural network is used for approximating the inversion of the industrial robot to implement decoupling and linearization of the industrial robot. Secondly, the composite pseudo linear system, which is composed of the ANN-inversion system and the controlled industrial robot, is equivalent to a linear system with disturbance in view of the uncertainties of the industrial robot and the approximation error of the BP neural network. Then, two FOSMCs are designed respectively based on the SMC theory and fractional calculus for the two subsystems, and the stability analysis is given. Finally, case study is fulfilled under different conditions, and results show the effectiveness of the proposed control scheme.