A robust visual servo control system for narrow seam double head welding robot

In this paper, an image-based visual servo control system is developed and integrated into a double head welding robot for CO2 gas shielded arc welding, which is mainly used to weld the narrow seam whose width is generally less than 0.2 mm. Firstly, a robust and reliable image processing algorithm is presented to reliably extract the seam feature during the welding. A statistic-based filter method is proposed to effectively weaken the image noises caused by various disturbances. The region of interest (ROI) of the image is determined to improve the efficiency and reliability of the image processing. The feature line of the narrow seam is extracted by combining the technique of the least-square line fitting with the Hough transform method. Secondly, a visual servo control system is designed to realize accurate tracking and initial point localization of the narrow seam. Moreover, an image error filter and output pulse verification are utilized to guarantee the reliability and stability of the control system. Finally, a series of experiments with real welding robots are conducted in the production line of a container manufacturing factory, demonstrating the satisfactory performance and high welding quality of the proposed visual servo control system.