基于PGA-2-OPT算法的焊接机器人路径规划  

Path Planning of Welding Robot Based on PGA-2-OPT Algorithm

作　　者：刘良斌[1] 赵自鹏[1] 王瑶茜[1] LIU Liang-bin;ZHAO Zi-peng;WANG Yao-xi(Hunan Industry Polytechnic Department of Electrical Engineering,Changsha 410208,Hunan)

机构地区：[1]湖南工业职业技术学院电气工程学院,湖南长沙410208

出　　处：《电脑与电信》2023年第8期42-45,共4页Computer & Telecommunication

基　　金：湖南省自然科学基金资助项目,项目编号:2022JJ60035。

摘　　要：针对白车身焊接过程中焊点数量大、焊接顺序规划凭借经验的问题,提出了将焊接顺序和路径的规划问题转换为TSP问题的方案。传统遗传算法对种群多样化要求较高,后期易陷入局部最优问题;基本单亲遗传算法存在路径交叉问题。文章先对焊接任务采用单亲遗传算法得到相对最优解,然后采用2-OPT领域搜索策略,消除路径存在的交叉问题。文章以宝马某车型右前门板焊接任务为例,对焊接轨迹进行优化。改进后的车门焊接轨迹在同等进化代数的情况下,比传统遗传算法缩短了46%,比原有的单亲遗传算法缩短了29%。实验表明,通过PGA-2-OPT算法可以实现车门板焊接路径的优化,并能很好地消除路径交叉问题。In view of the large number of welding spots and the problems of welding sequence planning based on experience in the process of body in white welding,a scheme is proposed to transform the problem of welding sequence and path planning into TSP problem.The traditional genetic algorithm has high requirements on population diversity and is easy to fall into the local optimal problem in the late stage.Basic single parent genetic algorithm has the problem of path crossing.In this paper,the relative optimal solution of welding task is obtained by single parent genetic algorithm,and then the 2-OPT domain search strategy is adopted to eliminate the crossing problem of paths.Take the welding task of the right front door panel of a BMW model for example,this paper optimizes the welding trajectory.Under the same evolutionary algebra,the improved door welding trajectory is 46%shorter than the traditional genetic algorithm and 29%shorter than the original single parent genetic algorithm.Experiments show that PGA-2-OPT algorithm can be used to optimize the welding path of vehicle door panels and eliminate the problem of path crossing.

关 键 词：单亲遗传算法 白车身焊接 2-OPT 

分 类 号：TP24[自动化与计算机技术—检测技术与自动化装置]