Detection of weld position and seam tracking based on Kalman filtering of weld pool images

Abstract

This paper presents a novel method of detection of weld and seam tracking during gas tungsten arc welding (GTAW). The weld position equations are established based on an estimation of the centroid position of the weld pool images. A Kalman filter (KF) is applied to the weld pool images from a visual sensor to recursively compute the solution to the weld position equations. The estimation of the state provided by a Kalman filter can be used to estimate the weld feature location on the image plane at the next sampling time and its uncertainty. This allows adjusting the welding torch position in real time, which may sensibly reduce processing time and improve seam tracking accuracy. The effectiveness of the proposed algorithm in the presence of weld pool image noise has been tested by the computer simulations and actual welding experiments. Also, the results of experiments have demonstrated the robustness of weld position detection for seam tracking.