Neural Fuzzy Sliding Mode Control for a Robot Manipulator

To improve the performance of the sliding mode controller, a neural fuzzy sliding mode controller is proposed in this paper. This control strategy combines the neural control, fuzzy control and sliding mode control. The conventional sliding mode controller is divided into two parts: equivalent controller and corrective controller. To realize the control without the model of system, a neural controller is designed to estimate the equivalent controller. To diminish the chattering, a fuzzy controller is designed, whether requiring corrective controller or not is decided according to sliding surface by the fuzzy controller. The simulation studies have been carried out to show the tracking performance of two-links robot manipulator. The results indicate the superior performance of the proposed controller.