Experiments on Automatic Seam Detection for a MIG Welding Robot

To make robotic welding more flexible, vision systems are used to detect the weld seam and plan a path for the robot to follow. In this paper an image processing technique is introduced that can automatically detect the weld seam in a "butt-weld" configuration. This method is an improvement on the existing K-Cosine algorithm. The 3D location of weld points is determined using a robot Hand-in-Eye stereo vision system. This paper will also introduce a practical method for robot and Hand/Eye calibration. The validity of these methods will be verified through experiments using a MIG welding robot.