A Teaching-Free Welding Method Based on Laser Visual Sensing System in Robotic GMAW

Hou Zhen,Xu Yanling,Xiao Runquan,Chen Shanben
DOI: https://doi.org/10.1007/s00170-020-05774-0
IF: 3.563
2020-01-01
The International Journal of Advanced Manufacturing Technology
Abstract:At present, the majority of welding robots belong to the teach-and-playback category in welding manufacturing engineering applications; trajectory teaching in advance of welding is time-consuming and lack of efficiency. The currently published welding seam tracking methods are also based on existing trajectories. Intelligent Welding Manufacturing (IWM) is the current research focus in welding manufacturing. The core technology of IWM is Intelligent Robot Welding Technology (IRWT). A sensing-technology-based system was the key to realize IRWT. In this paper, a teaching-free welding method based on laser visual sensing system (LVSS) for robotic gas metal arc welding (GMAW) is studied. First, a LVSS was established. A fast-unified calibration method for LVSS is proposed to improve the calibration efficiency. Then, using an image processing method based on prior knowledge, feature point with sub-pixel accuracy can be obtained in real-time. Finally, an online welding trajectory planning method is proposed to implement teaching-free welding. In order to verify the accuracy and robustness of the proposed method, experiments on V-grooves and fillet welds were performed. The results showed that the control accuracy on the V-groove and the fillet welds is suitable for most robot welding applications.
What problem does this paper attempt to address?