Seam Tracking and Dynamic Process Control for High Precision Arc Welding

A local intelligent welding robot system with low-cost vision sensor was developed. Especially, adaptive control for the correction of welding path deviation, welding current and wire feeding velocity in real time were constructed. For the seam-tracking module, an improved adaptive threshold segmentation algorithm was applied to detect the welding curve. The results show that direct imaging with advanced image processing algorithm can provide precise and robust information of the weld groove. At the same time, a nonlinear self-corrected controller was constructed. Input parameters including geometric parameters of the weld pool were captured from dynamic image acquired by visual sensing system. In order to keep a sound weld appearance, the variation of the weld gap compensatory feed-back module was added into the control system. This local intelligent welding robot system was verified on the face flange of aluminum alloy.