Seam sensing of multi-layer and multi-pass welding based on grid structured laser

Multi-layer and multi-pass welding is widely used in some fields such as shipbuilding and boiler manufacturing. However, single line structured laser is insufficient to obtain the position and pose of the finished welding seams and predict next welding seam after based layer welding had been completed. In this paper, a new vision sensing method which uses grid structured laser for weld seam sensing is presented. It detects the previous weld beads to predict the following welding path position of the next layer. Based on image capturing and image processing algorithm, a high-quality image with distorted grid is obtained. Since extracting feature points from distorted grid structured laser is much more difficult than that from single line structured laser, a robust algorithm is proposed. Through calibrating and calculating of coordinate conversion, the position data of seams is determined in the world Cartesian coordinate system. The detecting and scanning experiments are ultimately performed, which prove that the seam sensing method is feasible and effective.