基于经验学习的空间机器人精细抓捕方法

doi:10.3969/j.issn.1674-1579.2015.05.001

摘要/Abstract

摘要： 机器人在空间环境下完成对复杂形状物体的精细操作，要求其具有较强的自主操作规划能力.从学习人类经验出发，模仿人类经验生成来建立经验数据库，提出有效的物体分类方法，在物体类别基础上，识别同类物体的可抓取区域，并提出“单拇指”姿态规划方法，将对物体的操作经验泛化到同类物体的操作规划中.试验证明了算法的有效性.

关键词: 精细操作, 空间机器人, 人类经验, 抓取规划, 物体分类

Abstract: It is a challenging problem for space robots autonomously to accomplish precise manipulation on objects with complex geometry. In this paper, an effective method for object classification and graspable components identification is firstly proposed by imitating the generation of human experience. Then, a “single” finger grasp planning method is developed which is able to generalize the grasp to the objects in the same category. Experiments verify the effectiveness of the proposed methods.

Key words: precise manipulation； space robot； human experience； grasp planning, object classification

中图分类号:

V448

孙富春, 刘春芳. 基于经验学习的空间机器人精细抓捕方法[J]. 空间控制技术与应用, 2015, 41(5): 1-5.

SUN Fu-Chun, LIU Chun-Fang. Space Robot Precise Manipulation Based on Experience Learning[J]. , 2015, 41(5): 1-5.

参考文献

［1］LI M, BEKIROGLU Y, KRAGIC D, et al. Learning of grasp adaptation through experience and tactile sensing［C］//IEEE/RSJ international conference on Intelligent Robots and Systems. Washington D.C.:IEEE, 2014:33393346. ［2］ELKHOURY S, SAHBANI A. A new strategy combining empirical and analytical approaches for grasping unknown 3d objects［J］. Robotics and Autonomous Systems, 2010, 58(5), 497507. ［3］DANG H, ALLEN P K. Grasp adjustment on novel objects using tactile experience from similar local geometry［C］//IEEE/RSJ international conference on Intelligent Robots and Systems. Washington D.C.:IEEE, 2013: 40074012. ［4］BOHG J, MORALES A, ASFOUR T, et al. Datadriven grasp synthesisa survey ［J］. IEEE Transaction Robot, 2014, 30(2), 289309. ［5］HUANG B, ELKHOURY S, LI M, et al. Learning a real time grasping strategy［C］//IEEE international conference on Robotics and Automation. Washington D.C.:IEEE, 2013: 593600. ［6］MADRY M, SONG D, KRAGIC D. From object categories to grasp transfer using probabilistic reasoning［C］//IEEE international conference on Robotics and Automation. Washington D.C.:IEEE, 2012: 17161723. ［7］NIKANDROVA E, KYRKI V. Categorybased task specific grasping［J］. Robotics and Autonomous Systems, 2015, 70: 2535. ［8］HILLENBRAND U, ROA M. Transferring functional grasps through contact warping and local replanning［C］//IEEE/RSJ international conference on Intelligent Robots and Systems. Washington D.C.:IEEE, 2012: 29632970. ［9］GRANA C, DAVOLIO M, CUCCHIARA R. Similaritybased retrieval with MPEG7 3d descriptors: performance evaluation on the Princeton shape benchmark［J］. Digital Libraries: Research and Development, First international DELOS Conference, 2007: 308317.

[1]	张文辉, 陈浩文, 闻志, 叶晓平. 面向未知目标的柔性关节空间机器人滑模控制[J]. 空间控制技术与应用, 2021, 47(3): 49-56.
[2]	冯钧, 孔建寿, 王刚. 一种柔性空间双臂机器人的协同控制和避障方法[J]. 空间控制技术与应用, 2020, 46(5): 10-17.
[3]	沈震, 罗璨, 商秀芹, 白天翔, 董西松, 宋晓光, 熊刚, 王飞跃. 空间平行机器与平行制造[J]. 空间控制技术与应用, 2019, 45(4): 80-.
[4]	胡晓东, 黄学祥, 胡天健, 王峰林, 梁书立. 一种动态环境下空间机器人的快速路径规划方法[J]. 空间控制技术与应用, 2018, 44(5): 14-21.
[5]	韩冬, 黄攀峰, 刘习尧. 空间机器人抓捕目标碰撞后的复合体稳定控制[J]. 空间控制技术与应用, 2018, 44(5): 38-46.
[6]	郑海潮, 何俊, 陈龙海. 基于导纳控制的空间精细操作半物理仿真[J]. 空间控制技术与应用, 2018, 44(4): 8-15.
[7]	王勇, 唐强, 徐拴锋, 朱志斌, 何英姿, 魏春岭. 串并混联空间机械臂全自由度轨迹规划算法[J]. 空间控制技术与应用, 2015, 41(5): 13-18.
[8]	刘冬雨, 刘宏, 黄龙飞. 基于Smith预测控制的空间机器人遥操作研究[J]. 空间控制技术与应用, 2015, 41(5): 19-.
[9]	王颖, 韩冬, 刘涛. 空间机器人协调控制全物理仿真设计与验证[J]. 空间控制技术与应用, 2015, 41(2): 30-35.
[10]	张涛, 陈章, 王学谦, 梁斌. 空间机器人遥操作关键技术综述与展望[J]. 空间控制技术与应用, 2014, 40(6): 1-9.
[11]	郑永洁, 张笃周, 谌颖. 空间机器人气浮式物理仿真系统有效性研究[J]. 空间控制技术与应用, 2010, 36(6): 33-38.
[12]	张军^1，2，胡海霞^1，2，邢琰^1，2. 空间机器人退步控制器设计[J]. 空间控制技术与应用, 2009, 35(1): 7-12.