面向3维物体的三指机械手“包笼抓取”方法  

Caging Grasp of 3D Objects with a Three-finger Gripper

作　　者：苏建华[1] 刘传凯[2] 王智伟 黄开启[3] SU Jianhua;LIU Chuankai;WANG Zhiwei;HUANG Kaiqi(Institute of Automation,Chinese Academy of Sciences,Beijing 100190,China;Beijing Aerospace Control Center,Beijing 100094,China;Jiangxi University of Science and Technology,Ganzhou 341000,China)

机构地区：[1]中国科学院自动化研究所,北京100190 [2]北京航天飞行控制中心,北京100094 [3]江西理工大学,江西赣州.341000

出　　处：《机器人》2021年第2期129-139,共11页Robot

基　　金：国家自然科学基金重大研究计划(91848109);山东省重大科技创新工程(2019JZZY010430);北京市自然科学基金(4182068)。

摘　　要：研究了面向3维多面体和类多面体的三指机械手包笼抓取策略.为了减小计算量,将高维包笼抓取构形的搜索问题转换成低维空间中包笼抓取集合的搜索问题,提出了利用机械手手指在构形空间中形成的约束域搜索包笼抓取构形的方法.这种方法搜索到的包笼抓取构形能够收敛到不完全形封闭抓取构形.最后,通过Barrett手的抓取实验验证了算法的有效性.The caging grasps of 3 D polyhedron and polyhedron-like objects with a three-finger gripper are studied. For the detection problem of grasp configuration, the high-dimensional cage is converted into several lower-dimensional cage sets, to reduce the computation cost. A detection method of the caging grasp configuration is proposed based on the constraint region formed by the gripper fingers in the configuration space. The caging grasp configurations detected by the proposed method can converge to a partial form-closure grasp configuration. Finally, the effectiveness of the proposed method is demonstrated by experiments on a Barrett hand.

关 键 词：包笼 抓取 约束域 三指机械手 

分 类 号：TP24[自动化与计算机技术—检测技术与自动化装置]