A new seam-tracking algorithm through characteristic-point detection for a portable welding robot

The welding task in double-hulled structures in shipyards and in steel-frame structures is hazardous and difficult due to the toxic gas and limited workspace. Therefore, many efforts have been undertaken for automation. The main challenge for automation is the development of a simple and robust seamtracking algorithm that can be applied to a portable welding robot that operates under irregular and diverse task conditions in the workspace. We developed a seam-tracking algorithm for weaving weld path planning using a laser displacement sensor. The goal of the proposed algorithm is to detect the seam of single-butt welding with manually tack-welded non-zero gaps. The focus is on keeping the algorithm simple and affordable so that it can be applied to portable robots that operate in hazardous fields. The algorithm consists of four steps: scanning, filtering, generation of the reference points, and path planning. In the scanning process, the depth data of a cross-section of the seam profile is obtained. Next, a Gaussian filter is used to remove noise from the raw data. A differential characteristic-point detection algorithm is applied to the filtered data to detect the reference points that represent the shape and location of the gap to be welded. Finally, path planning for single-V butt multi-pass welding is done based on the detected reference points. A portable four-axis welding robot is built using the developed algorithm. The algorithm is validated through welding experiments regarding a single-V butt welding task with a manually tack-welded non-zero gap. Crown Copyright & 2011 Published by Elsevier Ltd. All rights reserved.