Visual-based Intelligent Control System for Robotic Gas Metal Arc Welding

Abstract

Sensing and control the weld pool is a crucial problem for robotic gas-metal arc welding (GMAW) process. In present research, a special vision sensing system, assisted by a narrow-band filter which could overcome the influence of the strong arc light of GMAW, was integrated with the morphological image-processing algorithm to improve the weld pool edge extracting performance. On the basis of welding pool image vision sensing and processing system, an intelligent controller based on fuzzy and expert system (ES) for robotic GMAW process is proposed. The experiments show that the real-time and precision requirements for detecting and control of weld pool width of robotic GMAW process could be satisfied by the established system.