面向车身数字化工艺开发的机器人动态碰撞检测算法研究  

Research on Robot Dynamic Collision Detection Algorithm for Body Digital Process Development

作　　者：王雪琪 赵文政 刘银华[1] 杨玉芳 李晓甜 WANG Xueqi;ZHAO Wenzheng;YANG Yufang;LI Xiaotian;LIU Yinhua

机构地区：[1]上海理工大学,上海200093 [2]上汽通用汽车有限公司,上海200120

出　　处：《上海汽车》2022年第3期55-62,共8页Shanghai Auto

基　　金：国家自然科学基金面上项目(51875362);机械系统与振动国家重点实验室开放基金(MSV202010);上汽通用汽车有限公司的产学研项目支持

摘　　要：在汽车行业制造流程快速向“数字化”与“智能化”转型的背景下,制造工艺自动开发技术获得学术与产业界的广泛重视。文章针对车身制造过程多机器人工位中自动化工艺路径的规划问题,提出基于智能化算法的多机器人系统动态碰撞检测与协调路径规划算法。尤其针对动态碰撞检测算法中存在的“隧道效应”导致的计算效率低和准确性不足问题,提出基于分级连续碰撞检测的GJK算法,实现机器人与机器人、机器人与车身之间的动态碰撞检测。最后,以搭载视觉传感器的多机器人检测工位为例开展案例应用与方法验证,说明了所提出算法的有效性,为汽车行业数字化工艺路径自动开发提供理论依据。Under the background of the rapid transformation of the manufacturing process of the automobile industry to “digitalization”and “intelligence”,the automatic development technology of manufacturing process has received extensive attention from academy and industries. Aiming at the problem of automated process path planning in multi-robot stations in the car body manufacturing process,a dynamic collision detection and coordination path planning algorithm are proposed for multirobot systems based on intelligent algorithms. In particular,in view of the low computational efficiency and insufficient accuracy caused by the “tunnel effect”in the dynamic collision detection algorithm,the GJK algorithm based on hierarchical continuous collision detection is proposed to realize dynamic collision detection between robots and robots and between robots and car bodies. Finally,taking a multi-robot inspection station equipped with a vision sensor as an example,a case application and method verification are carried out to illustrate the effectiveness of the proposed algorithm and provide a theoretical basis for the automatic development of digital process paths in the automobile industry.

关 键 词：质量检测 智能制造 多机器人系统 碰撞检测 车身 

分 类 号：U468.22[机械工程—车辆工程] TP242[交通运输工程—载运工具运用工程]