Robot Welding Path Planning and Application Based on Graphical Computing

Abstract

As a classic problem in artificial intelligence research, robot welding path planning has been extensively studied. Related scholars have also proposed many solutions, such as heuristic algorithm, neural network, genetic simulated annealing algorithm, improved genetic algorithm. But there are still deficiencies in welding torch posture, welding position, and robot motion stability. Because of the characteristics of the welding seam have a vital influence on the path planning of the welding robot, it is also the basis for ensuring the welding accuracy. From the perspective of graphic calculation, the graphical computing method of precise welds is analyzed, the point cloud graphics of the welded parts are used to calculate the overlap of primitives, and the accurate and rapid extraction of weld features are realized by changing the graphic representation of the welded parts model. According to the connection characteristics of the weldment, the characteristics of the weldment are collected, and a simple, fast, and more versatile method for extracting and calculating weld features is designed, and the weld features are discretized. Discrete weld feature points are used as the basis for path planning of the welding robot to carry out reasonable welding path planning, which reduces the manual teaching process and workload. Finally, a robot welding path planning method based on graphical computing is proposed, and corresponding simulation experiments are carried out.