File Download
There are no files associated with this item.
Links for fulltext
(May Require Subscription)
- Publisher Website: 10.1109/LRA.2021.3060636
- Scopus: eid_2-s2.0-85101736179
- WOS: WOS:000629028400033
- Find via
Supplementary
- Citations:
- Appears in Collections:
Article: Living Object Grasping Using Two-Stage Graph Reinforcement Learning
Title | Living Object Grasping Using Two-Stage Graph Reinforcement Learning |
---|---|
Authors | |
Keywords | Deep learning in grasping and manipulation dexterous manipulation grasping in-hand manipulation reinforcement learning |
Issue Date | 2021 |
Publisher | Institute of Electrical and Electronics Engineers. The Journal's web site is located at https://www.ieee.org/membership-catalog/productdetail/showProductDetailPage.html?product=PER481-ELE |
Citation | IEEE Robotics and Automation Letters, 2021, v. 6 n. 2, p. 1950-1957 How to Cite? |
Abstract | Living objects are hard to grasp because they can actively dodge and struggle by writhing or deforming while or even prior to being contacted and modeling or predicting their responses to grasping is extremely difficult. This letter presents an algorithm based on reinforcement learning (RL) to attack this challenging problem. Considering the complexity of living object grasping, we divide the whole task into pre-grasp and in-hand stages and let the algorithm switch between the stages automatically. The pre-grasp stage is aimed at finding a good pose of a robot hand approaching a living object for performing a grasp. Dense reward functions are proposed for facilitating the learning of right hand actions based on the poses of both hand and object. Since an object held in hand may struggle to escape, the robot hand needs to adjust its configuration and respond correctly to the object's movement. Hence, the goal of the in-hand stage is to determine an appropriate adjustment of finger configuration in order for the robot hand to keep holding the object. At this stage, we treat the robot hand as a graph and use the graph convolutional network (GCN) to determine the hand action. We test our algorithm with both simulation and real experiments, which show its good performance in living object grasping. More results are available on our website: https://sites.google.com/view/graph-rl. |
Persistent Identifier | http://hdl.handle.net/10722/300572 |
ISSN | 2023 Impact Factor: 4.6 2023 SCImago Journal Rankings: 2.119 |
ISI Accession Number ID |
DC Field | Value | Language |
---|---|---|
dc.contributor.author | Hu, Z | - |
dc.contributor.author | Zheng, Y | - |
dc.contributor.author | Pan, J | - |
dc.date.accessioned | 2021-06-18T14:53:55Z | - |
dc.date.available | 2021-06-18T14:53:55Z | - |
dc.date.issued | 2021 | - |
dc.identifier.citation | IEEE Robotics and Automation Letters, 2021, v. 6 n. 2, p. 1950-1957 | - |
dc.identifier.issn | 2377-3766 | - |
dc.identifier.uri | http://hdl.handle.net/10722/300572 | - |
dc.description.abstract | Living objects are hard to grasp because they can actively dodge and struggle by writhing or deforming while or even prior to being contacted and modeling or predicting their responses to grasping is extremely difficult. This letter presents an algorithm based on reinforcement learning (RL) to attack this challenging problem. Considering the complexity of living object grasping, we divide the whole task into pre-grasp and in-hand stages and let the algorithm switch between the stages automatically. The pre-grasp stage is aimed at finding a good pose of a robot hand approaching a living object for performing a grasp. Dense reward functions are proposed for facilitating the learning of right hand actions based on the poses of both hand and object. Since an object held in hand may struggle to escape, the robot hand needs to adjust its configuration and respond correctly to the object's movement. Hence, the goal of the in-hand stage is to determine an appropriate adjustment of finger configuration in order for the robot hand to keep holding the object. At this stage, we treat the robot hand as a graph and use the graph convolutional network (GCN) to determine the hand action. We test our algorithm with both simulation and real experiments, which show its good performance in living object grasping. More results are available on our website: https://sites.google.com/view/graph-rl. | - |
dc.language | eng | - |
dc.publisher | Institute of Electrical and Electronics Engineers. The Journal's web site is located at https://www.ieee.org/membership-catalog/productdetail/showProductDetailPage.html?product=PER481-ELE | - |
dc.relation.ispartof | IEEE Robotics and Automation Letters | - |
dc.rights | IEEE Robotics and Automation Letters. Copyright © Institute of Electrical and Electronics Engineers. | - |
dc.rights | ©20xx IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other uses, in any current or future media, including reprinting/republishing this material for advertising or promotional purposes, creating new collective works, for resale or redistribution to servers or lists, or reuse of any copyrighted component of this work in other works. | - |
dc.subject | Deep learning in grasping and manipulation | - |
dc.subject | dexterous manipulation | - |
dc.subject | grasping | - |
dc.subject | in-hand manipulation | - |
dc.subject | reinforcement learning | - |
dc.title | Living Object Grasping Using Two-Stage Graph Reinforcement Learning | - |
dc.type | Article | - |
dc.identifier.email | Pan, J: jpan@cs.hku.hk | - |
dc.identifier.authority | Pan, J=rp01984 | - |
dc.description.nature | link_to_subscribed_fulltext | - |
dc.identifier.doi | 10.1109/LRA.2021.3060636 | - |
dc.identifier.scopus | eid_2-s2.0-85101736179 | - |
dc.identifier.hkuros | 323044 | - |
dc.identifier.volume | 6 | - |
dc.identifier.issue | 2 | - |
dc.identifier.spage | 1950 | - |
dc.identifier.epage | 1957 | - |
dc.identifier.isi | WOS:000629028400033 | - |
dc.publisher.place | United States | - |