Research on trajectory tracking of a parallel robot based on neural network PID control

Parallel robots have very good performance in terms of rigidity, strength-to-weight ratios and positioning accuracy. A 2-DOF redundantly actuated parallel robot is taken as the object of study in this paper. As far as trajectory tracking performance, controller is essential. In order to improve the trajectory tracking performance, a neural network PID (proportion integral differential) controller is adopted. The backpropagation (BP) neural network is used to adjust the PID parameters. A simple proportion differential (PD) controller is selected to compare with the proposed controller. A circular motion control of the end-effector is studied using the two control methods. Simulation results show that the neural network PID controller is more effective than the PD controller in terms of reducing tracking errors, and it is a practical and valid alterative to parallel robots.