A point cloud-based welding trajectory planning method for plane welds

Plane welds are a common type of weld in industrial sites. When the welding robot multiple types plane at same time, traditional teaching and programming modes become relatively complicated. Therefore, order to solve problem automatic various welds, taking V-type butt, I-type lap as examples, this paper proposes extraction method based on 3D vision. Firstly, realize line segmentation workpiece point cloud, we establish concept cloud density. Secondly, segment edge lines, an iterative algorithm RANSAC is proposed. Then, geometric features workpiece, for extracting feature points centroid positioning Finally, least squares used fit seam complete trajectory planning. The experimental results show that can well planning simultaneously without programming.