Abstract: Because of the low absolute positioning accuracy, industrial robot is difficult to be used in aerospace high precision manufacturing. There are many factors that affect the positioning error of robot, which poses a severe challenge to the accurate establishment of its error model. The existing modeling methods usually associate the robot's positioning error with its pose, ignoring the influence of joint backlash on its positioning error in the same pose. In order to improve the absolute positioning accuracy of industrial robot, an error similarity accuracy compensation method for industrial robot considering joint backlash is proposed. Based on the improved Denavit-Hartenberg model, a comprehensive identification model including robot geometric error, coordinate system error and transmission error is established, and the joint backlash is identified by using the least square method. According to the identified parameters such as joint backlash, the error similarity model is constructed. Three types of robots are used to verify the feasibility and universality of the method for improving the absolute positioning accuracy of the robot. Finally, the drilling experiment is conducted by KUKA KR500-3 robot. The experimental results show that, compared with the traditional method, the positioning error of the robot is reduced by about 0.1 mm, the accuracy is improved by more than 30%, and the hole position accuracy is improved from 0.701 mm to 0.134 mm. From these results it is concluded that this method provides a technical means to effectively improve the absolute positioning accuracy of industrial robot.

Key words: industrial robot, backlash, error similarity, parameter identification, error compensation