The tracking method of robot arm trajectory based on artificial neural network evolution

摘要

Artificial neural network evolutionary method is a new machine learning method. Aiming at the control problems of high time-varying, nonlinear and strong coupling of robot arm, a nonlinear control method based on neural network evolutionary approximation is proposed. Based on the kinematics calculation method of n-joint controlled object, a RBF neural network controller is designed to track and approximate the target trajectory. The stability and feasibility of the system are demonstrated by using the integral Lyapunov method. Taking the trajectory of a two-joint manipulator as an example of MATLAB simulation, the results show that the method can effectively reduce the system modeling error, accurately track the trajectory of the manipulator, and improve the accuracy of system control.