中国科学院机构知识库网格
Chinese Academy of Sciences Institutional Repositories Grid
首页 机构 成果 学者
  
  1. CAS IR Grid
  2. 自动化研究所
  3. 中国科学院自动化研究所
  4. 多模态人工智能系统全国重点实验室
  5. 机器人应用与理论组
Learning Sliding Policy of Flat Multi-target Objects in Clutter Scenes

文献类型：期刊论文

作者Wu, Liangdong1; Wu, Jiaxi2; Li, Zhengwei3; Chen, Yurou2; Liu, Zhiyong1,2,4
刊名INFORMATION TECHNOLOGY AND CONTROL
出版日期2024
卷号53期号:1页码:5-18
关键词Deep Learning in Manipulation Reinforcement Learning Robot Control Intelligent system sliding policy
ISSN号1392-124X
DOI10.5755/j01.itc.53.1.34708
通讯作者Liu, Zhiyong(zhiyong.liu@ia.ac.cn)
英文摘要In clutter scenes, one or several targets need to be obtained, which is hard for robot manipulation task. Especially, when the targets are flat objects like book, plates, due to limitation of common robot end-effectors, it will be more challenging. By employing pre-grasp operation like sliding, it becomes feasible to rearrange objects and shift the target towards table edge, enabling the robot to grasp it from a lateral perspective. In this paper, the proposed method transfers the task into a Parameterized Action Markov Decision Process to solve the problem, which is based on deep reinforcement learning. The mask images are taken as one of observations to the network for avoiding the impact of noise of original image. In order to improve data utilization, the policy network predicts the parameters for the sliding primitive of each object, which is weight-sharing, and then the Q-network selects the optimal execution target. Meanwhile, extra reward mechanism is adopted for improving the efficiency of task actions to cope with multiple targets. In addition, an adaptive policy scaling algorithm is proposed to improve the speed and adaptability of policy training. In both simulation and real system, our method achieves a higher task success rate and requires fewer actions to accomplish the flat multi-target sliding manipulation task within clutter scene, which verifies the effectiveness of ours.
资助项目National Key Research and Development Plan of China[2020AAA0108902] ; Strategic Priority Research Program of Chinese Academy of Science[XDB32050100] ; Dongguan Core Technology Research Frontier Project, China[2019622101001]
WOS研究方向Automation & Control Systems ; Computer Science
语种英语
WOS记录号WOS:001280512700001
出版者KAUNAS UNIV TECHNOLOGY
资助机构National Key Research and Development Plan of China ; Strategic Priority Research Program of Chinese Academy of Science ; Dongguan Core Technology Research Frontier Project, China
源URL[http://ir.ia.ac.cn/handle/173211/59403]  
专题自动化研究所_复杂系统管理与控制国家重点实验室_机器人应用与理论组
通讯作者Liu, Zhiyong
作者单位1.Univ Chinese Acad Sci, Sch Artificial Intelligence, Beijing, Peoples R China
2.Chinese Acad Sci, Inst Automat, Beijing, Peoples R China
3.Univ Chinese Acad Sci, Sch Artificial Intelligence, Beijing, Peoples R China
4.Chinese Acad Sci, Cloud Comp Ctr, Dongguan, Guangdong, Peoples R China
推荐引用方式
GB/T 7714		 Wu, Liangdong,Wu, Jiaxi,Li, Zhengwei,et al. Learning Sliding Policy of Flat Multi-target Objects in Clutter Scenes[J]. INFORMATION TECHNOLOGY AND CONTROL,2024,53(1):5-18.
APA Wu, Liangdong,Wu, Jiaxi,Li, Zhengwei,Chen, Yurou,&Liu, Zhiyong.(2024).Learning Sliding Policy of Flat Multi-target Objects in Clutter Scenes.INFORMATION TECHNOLOGY AND CONTROL,53(1),5-18.
MLA Wu, Liangdong,et al."Learning Sliding Policy of Flat Multi-target Objects in Clutter Scenes".INFORMATION TECHNOLOGY AND CONTROL 53.1(2024):5-18.

入库方式： OAI收割

来源：自动化研究所

浏览0
下载0
收藏0
其他版本

除非特别说明，本系统中所有内容都受版权保护，并保留所有权利。