Abstract
The nut projection welding process is one of the key technologies in industrial manufacturing. The teaching-reproduction robotic welding mode is usually deployed in a structured environment, where the robot performs simple and fixed repetitive operations through offline teaching. This method is difficult to meet flexible welding requirements for multiple types of objects, and requires high consistency in the manufacture of both tools and industrial parts. Based on the robot hand-eye vision system, a fully automatic nut projection welding robotic system is designed and developed in this paper. The system is mainly composed of a flexible robot pick-and-place system for stacked industrial parts and a high-precision robotic welding system to complete the entire process. A 6D pose estimation algorithm for workpiece image recognition in stacking scenes and a welding error calibration method based on hand-eye visual feedback are realized. The effectiveness and reliability of the system are verified by the actual production experiments.
Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
