DBO-BP based Positioning Error Compensation Method for Industrial Robots
In order to improve the absolute positioning accuracy of industrial robots,a method based on DBO-BP and offline feedforward correction was proposed.This method is suitable for the research on positioning error compensation of industrial robots.By using Latin Hypercube Sampling method to obtain the pose samples of industrial robots,and using BP neural network to establish an error prediction model,the DBO optimization algorithm was applied to improve the local optimal phenomenon,thus improving the convergence and robustness of the model.After offline feedforward compensation processing,the positioning error of industrial robots was reduced,and the absolute positioning accuracy of robots was greatly improved.This method can effectively improve the accuracy and stability of robots,and provides a feasible solution for the precise positioning problem of industrial robots.
NETL
NSTL
万方数据
