Fuzzy Support Vector Machines Control for 6-DOF Parallel Robot

Abstract

In order to realize the trajectory tracking control of six degrees of freedom parallel robot, the dynamics equation of six degrees of freedom parallel robot was established. The parallel robot has obvious nonlinear, uncertainty characteristics and external disturbance, so the sliding mode variable structure theory was introduced into the system control. A fuzzy support vector machines control strategy based on sliding mode control was designed to reduce the oscillation of the sliding mode control. Parameters of fuzzy support vector machines controller were optimized by hybrid learning algorithm, which combines least square algorithm with improved genetic algorithm, to get the optimal control performance for the controlled object. The controller designed consists of a fuzzy sliding mode controller and a fuzzy support vector machines controller, and the compensation controller is selected by comparing switching function with the thickness of boundary layer. Simulation results show that under the condition of model error and external disturbance, the control strategy designed gets tracking effect with high precision and speed.