Welding robotic hand-eye calibration method based on structured light plane

Abstract

The calibration of the relationship between the welding robot hand and the camera is an important issue in the process of industrial welding intelligentization. For the welding robot with end-mounted line structured light, this paper proposes a novel method of robot hand-eye calibration based on structured light plane. Firstly, move the the position of robot to make structured light plane intersect with the marker points multiple times and the image coordinates of the marker points are obtained. Then construct and solve linear equations to obtain the least squares solution of the calibration matrix. Finally, use the algorithm of integrated learning method bagging to obtain the average of the least squares solution. Compared with the traditional complicated calibration method, this method doesn't need a specific standard workpiece or calibration board which makes the calibration process more convenient. The results of experiment show that the calibration three-axis measurement accuracy is within ±0.20mm, which meet the requirements of industrial field welding precision.