[1]南文虎,徐付民,叶伯生.记忆推理的放射源抓取机器人运动规划[J].深圳大学学报理工版,2022,39(3):343-348.[doi:10.3724/SP.J.1249.2022.03343]
 NAN Wenhu,XU Fumin,and YE Bosheng.Research on motion planning of radioactive source grasping robot based on memory reasoning[J].Journal of Shenzhen University Science and Engineering,2022,39(3):343-348.[doi:10.3724/SP.J.1249.2022.03343]
点击复制

记忆推理的放射源抓取机器人运动规划
分享到:

《深圳大学学报理工版》[ISSN:1000-2618/CN:44-1401/N]

卷:
第39卷
期数:
2022年第3期
页码:
343-348
栏目:
电子与信息科学
出版日期:
2022-05-16

文章信息/Info

Title:
Research on motion planning of radioactive source grasping robot based on memory reasoning
文章编号:
202203013
作者:
南文虎徐付民叶伯生
1)兰州理工大学机电工程学院,甘肃兰州730050;2)华中科技大学国家数控工程中心,湖北武汉 430074
Author(s):
NAN Wenhu XU Fumin and YE Bosheng
1) School of Mechanical and Electrical Engineering, Lanzhou University of Technology, Lanzhou 730050,Gansu Province, P. R. China 2) The National CNC Engineering Center,Huazhong University of Science and Technology,Wuhan 430074,Hubei Province, P. R. China
关键词:
智能机器人 力觉反馈 历史抓取数据 记忆推理决策 自主抓取 Gazebo仿真器 放射流抓取
Keywords:
intelligent robot force feedback historical captured data memory reasoning decision autonomous grasping Gazebo simulator capture of radioactive sources
分类号:
TP24
DOI:
10.3724/SP.J.1249.2022.03343
文献标志码:
A
摘要:
针对机器人抓取铅罐内部放射源时,因铅罐的半封闭和强辐射环境导致机器视觉难以应用于放射源抓取的问题,提出一种记忆推理的强化学习抓取方法.基于机器视觉构建智能机器人抓取系统运动学模型;采用力觉反馈实现智能机器人与铅罐内部环境的交互;通过对历史抓取数据的记忆推理决策,实现对放射源的自主抓取.在机器人操作系统中使用Gazebo仿真器,分别采用蒙特卡罗采样法和基于记忆推理的强化学习抓取方法进行仿真,结果表明,基于记忆推理的强化学习抓取方法的平均抓取效率比蒙特卡罗采样法高84.67%,能高效地解决铅罐内部放射源的自主抓取问题.
Abstract:
Aiming at the problem that machine vision is difficult to be applied to the capture of radioactive sources due to the semi-closed and strong radiation environment of lead cans, we propose a memory reasoning reinforcement learning capture method. The kinematics model of intelligent robot capture system is constructed based on machine vision. The interaction between the intelligent robot and the internal environment of the lead is realized by force feedback. Through the memory reasoning decision of historical captured data, the autonomous capture of radioactive sources is realized. Using Gazebo simulator in robot operating system (ROS), Monte Carlo sampling method and reinforcement learning capture method based on memory reasoning are simulated, respectively. The results show that the reinforcement learning capture method based on memory reasoning, The average grasping efficiency is 84.67% higher than that of Monte Carlo sampling method, which can effectively solve the problem of autonomous grasping of radioactive sources in lead cans.

相似文献/References:

[1]黄海明,吴林源,林俊豪,等.基于内嵌型光学弯曲传感器的软体手感知[J].深圳大学学报理工版,2019,36(3):237.[doi:10.3724/SP.J.1249.2019.03229]
 HUANG Haiming,WU Linyuan,LIN Junhao,et al.Soft hand perception based on embedded optical fiber bending sensor[J].Journal of Shenzhen University Science and Engineering,2019,36(3):237.[doi:10.3724/SP.J.1249.2019.03229]
[2]夏德龙,吴耀华,王艳艳,等.基于智能机器人的“货到人”系统订单排序优化[J].深圳大学学报理工版,2019,36(6):696.[doi:10.3724/SP.J.1249.2019.06696]
 XIA Delong,WU Yaohua,WANG Yanyan,et al.Order sequence optimization for parts-to-picker intelligent robot system[J].Journal of Shenzhen University Science and Engineering,2019,36(3):696.[doi:10.3724/SP.J.1249.2019.06696]

更新日期/Last Update: 2022-05-30