RBF neural network PID trajectory tracking based on 6-PSS parallel robot

Abstract

Aiming at the problems of trajectory tracking of six-degree-of-freedom parallel robots, this paper proposes a control strategy combining RBF neural network and traditional PID. According to the nonlinear characteristics of parallel robot, RBF neural network is used to approximate the characteristics of arbitrary nonlinear function. The RBF neural network is optimized by gradient descent algorithm. The neural network PID controller is designed to find the optimal PID parameters through the adaptive self-learning ability of the neural network, thus achieving better trajectory tracking effect.