Abstract: In order to improve the absolute positioning accuracy of industrial robots, a method based on DBO-BP and offline feedforward correction is 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 is 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 is reduced, and the absolute positioning accuracy of robots is 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.
刘麒, 谭丁诚, 刘振刚, 王影. 基于DBO-BP的工业机器人定位误差补偿方法[J]. 吉林化工学院学报, 2024, 41(1): 59-66.
LIU qi , TAN Dingcheng , WANG ying. DBO-BP based Positioning Error Compensation Method for Industrial Robots . Journal of Jilin Institute of Chemical Technology, 2024, 41(1): 59-66.
