ArrayBot: Reinforcement Learning for Generalizable Distributed Manipulation through Touch
| Title: | ArrayBot: Reinforcement Learning for Generalizable Distributed Manipulation through Touch |
|---|---|
| Authors: | Xue, Zhengrong, Zhang, Han, Cheng, Jingwen, He, Zhengmao, Ju, Yuanchen, Lin, Changyi, Zhang, Gu, Xu, Huazhe |
| Publication Year: | 2023 |
| Collection: | Computer Science |
| Subject Terms: | Computer Science - Robotics, Computer Science - Machine Learning |
| More Details: | We present ArrayBot, a distributed manipulation system consisting of a $16 \times 16$ array of vertically sliding pillars integrated with tactile sensors, which can simultaneously support, perceive, and manipulate the tabletop objects. Towards generalizable distributed manipulation, we leverage reinforcement learning (RL) algorithms for the automatic discovery of control policies. In the face of the massively redundant actions, we propose to reshape the action space by considering the spatially local action patch and the low-frequency actions in the frequency domain. With this reshaped action space, we train RL agents that can relocate diverse objects through tactile observations only. Surprisingly, we find that the discovered policy can not only generalize to unseen object shapes in the simulator but also transfer to the physical robot without any domain randomization. Leveraging the deployed policy, we present abundant real-world manipulation tasks, illustrating the vast potential of RL on ArrayBot for distributed manipulation. Comment: ICRA24 |
| Document Type: | Working Paper |
| Access URL: | http://arxiv.org/abs/2306.16857 |
| Accession Number: | edsarx.2306.16857 |
| Database: | arXiv |
| FullText | Text: Availability: 0 CustomLinks: – Url: http://arxiv.org/abs/2306.16857 Name: EDS - Arxiv Category: fullText Text: View this record from Arxiv MouseOverText: View this record from Arxiv – Url: https://resolver.ebsco.com/c/xy5jbn/result?sid=EBSCO:edsarx&genre=article&issn=&ISBN=&volume=&issue=&date=20230629&spage=&pages=&title=ArrayBot: Reinforcement Learning for Generalizable Distributed Manipulation through Touch&atitle=ArrayBot%3A%20Reinforcement%20Learning%20for%20Generalizable%20Distributed%20Manipulation%20through%20Touch&aulast=Xue%2C%20Zhengrong&id=DOI: Name: Full Text Finder (for New FTF UI) (s8985755) Category: fullText Text: Find It @ SCU Libraries MouseOverText: Find It @ SCU Libraries |
|---|---|
| Header | DbId: edsarx DbLabel: arXiv An: edsarx.2306.16857 RelevancyScore: 1057 AccessLevel: 3 PubType: Report PubTypeId: report PreciseRelevancyScore: 1057.34582519531 |
| IllustrationInfo | |
| Items | – Name: Title Label: Title Group: Ti Data: ArrayBot: Reinforcement Learning for Generalizable Distributed Manipulation through Touch – Name: Author Label: Authors Group: Au Data: <searchLink fieldCode="AR" term="%22Xue%2C+Zhengrong%22">Xue, Zhengrong</searchLink><br /><searchLink fieldCode="AR" term="%22Zhang%2C+Han%22">Zhang, Han</searchLink><br /><searchLink fieldCode="AR" term="%22Cheng%2C+Jingwen%22">Cheng, Jingwen</searchLink><br /><searchLink fieldCode="AR" term="%22He%2C+Zhengmao%22">He, Zhengmao</searchLink><br /><searchLink fieldCode="AR" term="%22Ju%2C+Yuanchen%22">Ju, Yuanchen</searchLink><br /><searchLink fieldCode="AR" term="%22Lin%2C+Changyi%22">Lin, Changyi</searchLink><br /><searchLink fieldCode="AR" term="%22Zhang%2C+Gu%22">Zhang, Gu</searchLink><br /><searchLink fieldCode="AR" term="%22Xu%2C+Huazhe%22">Xu, Huazhe</searchLink> – Name: DatePubCY Label: Publication Year Group: Date Data: 2023 – Name: Subset Label: Collection Group: HoldingsInfo Data: Computer Science – Name: Subject Label: Subject Terms Group: Su Data: <searchLink fieldCode="DE" term="%22Computer+Science+-+Robotics%22">Computer Science - Robotics</searchLink><br /><searchLink fieldCode="DE" term="%22Computer+Science+-+Machine+Learning%22">Computer Science - Machine Learning</searchLink> – Name: Abstract Label: Description Group: Ab Data: We present ArrayBot, a distributed manipulation system consisting of a $16 \times 16$ array of vertically sliding pillars integrated with tactile sensors, which can simultaneously support, perceive, and manipulate the tabletop objects. Towards generalizable distributed manipulation, we leverage reinforcement learning (RL) algorithms for the automatic discovery of control policies. In the face of the massively redundant actions, we propose to reshape the action space by considering the spatially local action patch and the low-frequency actions in the frequency domain. With this reshaped action space, we train RL agents that can relocate diverse objects through tactile observations only. Surprisingly, we find that the discovered policy can not only generalize to unseen object shapes in the simulator but also transfer to the physical robot without any domain randomization. Leveraging the deployed policy, we present abundant real-world manipulation tasks, illustrating the vast potential of RL on ArrayBot for distributed manipulation.<br />Comment: ICRA24 – Name: TypeDocument Label: Document Type Group: TypDoc Data: Working Paper – Name: URL Label: Access URL Group: URL Data: <link linkTarget="URL" linkTerm="http://arxiv.org/abs/2306.16857" linkWindow="_blank">http://arxiv.org/abs/2306.16857</link> – Name: AN Label: Accession Number Group: ID Data: edsarx.2306.16857 |
| PLink | https://login.libproxy.scu.edu/login?url=https://search.ebscohost.com/login.aspx?direct=true&site=eds-live&scope=site&db=edsarx&AN=edsarx.2306.16857 |
| RecordInfo | BibRecord: BibEntity: Subjects: – SubjectFull: Computer Science - Robotics Type: general – SubjectFull: Computer Science - Machine Learning Type: general Titles: – TitleFull: ArrayBot: Reinforcement Learning for Generalizable Distributed Manipulation through Touch Type: main BibRelationships: HasContributorRelationships: – PersonEntity: Name: NameFull: Xue, Zhengrong – PersonEntity: Name: NameFull: Zhang, Han – PersonEntity: Name: NameFull: Cheng, Jingwen – PersonEntity: Name: NameFull: He, Zhengmao – PersonEntity: Name: NameFull: Ju, Yuanchen – PersonEntity: Name: NameFull: Lin, Changyi – PersonEntity: Name: NameFull: Zhang, Gu – PersonEntity: Name: NameFull: Xu, Huazhe IsPartOfRelationships: – BibEntity: Dates: – D: 29 M: 06 Type: published Y: 2023 |
| ResultId | 1 |