Research on automatic measurement method of end pose of binocular robot tool

Abstract

Aiming at the problems of long manual calibration time and large calibration error in robot hand-eye calibration process. Based on binocular measurement, this paper designs a hand-eye calibration method for robot based on depth camera. Using the aluminum block commonly used in production line as calibration object, the calibration object image is collected by depth camera, and the upper computer program is written based on OPENCV. The image is processed to obtain the coordinates of the reference points of the calibration objects in the visual coordinate system. Meanwhile, the coordinates in the basic coordinate system of the robot are obtained by the motion controller to calculate the hand-eye clearance of the robot. The hand-eye calibration of robot vision system is accomplished by using the transformation matrix. The experimental results show that the calibration error of this method is not higher than 0.40mm in plane and 0.46mm in space. The fixed time shall not exceed 5 minutes. The correctness and effectiveness of the method are verified and the efficiency of industrial robot calibration process is improved.