摘要
针对传统装配路径规划方法应用于复杂装配体时出现的组合爆炸问题,提出了一种基于人机交互的改进A-Star(A*)算法多层次装配路径规划方法。算法在传统启发式路径规划算法的基础上引入了干涉威胁概率、平滑度代价、权重系数参数,实现了算法不同侧重方向的最优路径寻找。算法首先根据基于人机交互的路径规划方法结合操作者的装配经验将复杂装配体划分为多个装配层次段,其次对各层次段中的装配零部件运用改进A*算法求解最优拆卸路径,并最终根据路径反演原则生成整个复杂装配体最优装配路径。结合算法仿真对比与KUKA工业机器人路径规划实例验证,得出上述方法较传统启发式路径规划方法提升了路径规划效率,满足了工业生产中复杂装配体自动路径规划需求。
A multi-level assembly path planning method based on improved A-Star(A*)algorithm with human-computer interaction is proposed to solve the problem of combinatorial explosion when traditional assembly path planning methods are applied to complex assemblies.Firstly,the traditional A*algorithm was improved by introducing some parameters,such as interference threat probability,smoothness cost,weight coefficient and so on,so as to realize the optimal path finding in different directions.Then,the human-machine interaction path planning method was used to divide the complex assembly into multiple levels,and the improved A*algorithm was used for the disassembly path planning of the components in each level.Through the path inversion,the entire assembly path was finally generated.The experimental simulation comparison and the six-degree-of-freedom robot assembly path planning experiment show that the efficiency of this method is 33.6%higher than the traditional assembly path planning method,which makes it possible to apply the automatic assembly path planning method to actual industrial production.
作者
李思良
袁庆霓
胡涞
黄鑫
LI Si-liang;YUAN Qing-ni;HU Lai;HUANG Xin(Key Laboratory of Advanced Manufacturing Technology,Ministry of Education,Guizhou Univ.,Guiyang Guizhou 550025,China)
出处
《计算机仿真》
北大核心
2020年第3期178-182,242,共6页
Computer Simulation
基金
基于大数据的虚拟仿真装配技术专利战略研究(黔科合基础[2017]1708)
基于分形几何的蜡染产品创新设计(黔科合支撑[2017]2870)
贵州省人力资源和社会保障厅贵州省留学人员科技创新项目(黔人项目资助合同(2016)06号)。
关键词
多层次装配
启发式算法
虚拟装配
路径规划
Multi-level assembly
Heuristic algorithm
Virtual assembly
Path planning
