Abstract: The welding process is one of the important links of vehicle manufacturing.The welding process planning of multi-robot involves the coupling of task allocation,multi-robot coordination and trajectory planning,which leads to the solution difficulties.A new task assignment method for coordinated motion planning of multi-welding robots was proposed to solve the problems such as narrow motion space and limited welding time of multi-robot in the welding process of complex structure of car body,and the conditions such as robot accessibility,no collision between robots and static environment and no conflict between multiple robots were comprehensively considered.The candidate trajectory library and trajectory conflict table were constructed.On this basis,an optimization model was established for welding task allocation and welding trajectory coordination planning.An improved hybrid leap-frog algorithm based on double matrix coding was proposed to optimize the solution.By adding two crossover operators,a mutation operator and a greedy placement strategy,the approximate optimal planning results of multi-robot welding tasks were obtained.A body in white welding case was used to verify the effectiveness of the proposed method.The results showed that the proposed method reduced the welding cycle of the station by 16.26% on average and improved the consistency of multi-robot running time by 30.87%.

Key words: welding process, task assignment, multiple constraints, coordinated motion planning

摘要： 焊接工艺是整车制造的重要环节之一,多机器人焊接工艺问题涉及任务分配、多机协调与运动规划等多个子问题,这些子问题之间的耦合关系导致工艺规划问题求解困难。针对车身复杂结构焊接过程中多机器人运动空间狭窄、焊接节拍受限等问题,综合考虑机器人的可达性、机器人与静态环境无碰与多机器人无冲突等条件,提出一种面向多焊接机器人协调运动规划的任务分配新方法。首先,构建候选轨迹库与轨迹冲突表,在此基础上建立面向焊点任务分配与焊接轨迹协调规划的优化模型,提出基于双矩阵编码的改进混合蛙跳算法进行优化求解。通过加入两种交叉算子、一种变异算子与贪婪放置策略,获得多机器人焊接任务近似最优规划结果。最后,通过某白车身焊接案例验证了所提方法有效性。同时,结果证明,工位焊接周期平均下降16.26%,多机器人运行时间一致性提升30.87%。

关键词: 焊接工艺, 任务分配, 多约束优化, 协调运动规划