Abstract: Welding robots have a wide range of applications in manufacturing industries. There are usually many welded joints in the welding task, and a reasonable welding path passes through these welded joints has a significant impact on the welding efficiency. Traditional manual path planning techniques can handle a small number of welded joints effectively, but when the number of welded joints is large, it is difficult to obtain the optimal path. The traditional manual path planning method is also time consuming and inefficient, and cannot guarantee optimality. Genetic algorithmparticle swarm optimization (GA-PSO) is based on the advantages of genetic algorithm (GA) and particle swarm optimization (PSO) to solve the welding robot path planning problem. The simulation results indicate that the algorithm has strong searching ability and practicality and is suitable for welding robot path planning.

Key words: Welding Robot, Path planning, Genetic algorithm, Particle swarm optimization, Global optimum

摘要： 焊接机器人在制造业中有广泛的应用。在焊接任务中通常有许多焊接接头，合理地规划焊接路径使其穿过这些焊接接头，对焊接效率的提高有积极的影响。传统的手工路径规划技术可以有效地处理少量焊接接头，但当焊接节点数目较大时，很难获得最优路径。传统的手工路径规划方法耗时长、效率低，不能保证最优。遗传粒子群优化算法（GA-PSO）基于遗传算法（GA）和粒子群优化算法（PSO）的优点来解决焊接机器人的路径规划问题。仿真结果表明，该算法具有较强的搜索能力和实用性，适用于焊接机器人路径规划。

关键词: 焊接机器人, 路径规划, 遗传算法, 粒子群优化算法, 全局最优