The Library
Reinforcement learning for robotic manipulation using simulated locomotion demonstrations
Tools
Tools
Tools
Kilinc, Ozsel and Montana, Giovanni (2022) Reinforcement learning for robotic manipulation using simulated locomotion demonstrations. Machine Learning, 111 . pp. 465-486. doi:10.1007/s10994-021-06116-1 ISSN 2632-2153.
|
PDF
WRAP-Reinforcement-learning-robotic-manipulation-locomotion-demonstrations-2021.pdf - Published Version - Requires a PDF viewer. Available under License Creative Commons Attribution 4.0. Download (2130Kb) | Preview |
Official URL: http://dx.doi.org/10.1007/s10994-021-06116-1
Abstract
Mastering robotic manipulation skills through reinforcement learning (RL) typically requires the design of shaped reward functions. Recent developments in this area have demonstrated that using sparse rewards, i.e. rewarding the agent only when the task has been successfully completed, can lead to better policies. However, state-action space exploration is more difficult in this case. Recent RL approaches to learning with sparse rewards have leveraged high-quality human demonstrations for the task, but these can be costly, time consuming or even impossible to obtain. In this paper, we propose a novel and effective approach that does not require human demonstrations. We observe that every robotic manipulation task could be seen as involving a locomotion task from the perspective of the object being manipulated, i.e. the object could learn how to reach a target state on its own. In order to exploit this idea, we introduce a framework whereby an object locomotion policy is initially obtained using a realistic physics simulator. This policy is then used to generate auxiliary rewards, called simulated locomotion demonstration rewards (SLDRs), which enable us to learn the robot manipulation policy. The proposed approach has been evaluated on 13 tasks of increasing complexity, and can achieve higher success rate and faster learning rates compared to alternative algorithms. SLDRs are especially beneficial for tasks like multi-object stacking and non-rigid object manipulation.
| Item Type: | Journal Article | ||||||||
|---|---|---|---|---|---|---|---|---|---|
| Subjects: | Q Science > Q Science (General) T Technology > TJ Mechanical engineering and machinery |
||||||||
| Divisions: | Faculty of Science, Engineering and Medicine > Engineering > WMG (Formerly the Warwick Manufacturing Group) | ||||||||
| Library of Congress Subject Headings (LCSH): | Reinforcement learning, Robotics , Robots -- Control systems | ||||||||
| Journal or Publication Title: | Machine Learning | ||||||||
| Publisher: | Springer | ||||||||
| ISSN: | 2632-2153 | ||||||||
| Official Date: | February 2022 | ||||||||
| Dates: |
|
||||||||
| Volume: | 111 | ||||||||
| Page Range: | pp. 465-486 | ||||||||
| DOI: | 10.1007/s10994-021-06116-1 | ||||||||
| Status: | Peer Reviewed | ||||||||
| Publication Status: | Published | ||||||||
| Access rights to Published version: | Open Access (Creative Commons) | ||||||||
| Date of first compliant deposit: | 22 December 2021 | ||||||||
| Date of first compliant Open Access: | 23 December 2021 |
Request changes or add full text files to a record
Repository staff actions (login required)
 |
View Item |
Downloads
Downloads per month over past year
