Accurate Vision-based Manipulation through Contact Reasoning
© 2020 IEEE. Planning contact interactions is one of the core challenges of many robotic tasks. Optimizing contact locations while taking dynamics into account is computationally costly and, in environments that are only partially observable, executing contact-based tasks often suffers from low accu...
Main Authors: | , , , |
---|---|
Other Authors: | |
Format: | Article |
Language: | English |
Published: |
Institute of Electrical and Electronics Engineers (IEEE)
2021
|
Online Access: | https://hdl.handle.net/1721.1/138353.2 |
_version_ | 1811091560723382272 |
---|---|
author | Bauza, Maria Wu, Jiajun Tenenbaum, Joshua B Rodriguez, Alberto |
author2 | Massachusetts Institute of Technology. Department of Brain and Cognitive Sciences. |
author_facet | Massachusetts Institute of Technology. Department of Brain and Cognitive Sciences. Bauza, Maria Wu, Jiajun Tenenbaum, Joshua B Rodriguez, Alberto |
author_sort | Bauza, Maria |
collection | MIT |
description | © 2020 IEEE. Planning contact interactions is one of the core challenges of many robotic tasks. Optimizing contact locations while taking dynamics into account is computationally costly and, in environments that are only partially observable, executing contact-based tasks often suffers from low accuracy. We present an approach that addresses these two challenges for the problem of vision-based manipulation. First, we propose to disentangle contact from motion optimization. Thereby, we improve planning efficiency by focusing computation on promising contact locations. Second, we use a hybrid approach for perception and state estimation that combines neural networks with a physically meaningful state representation. In simulation and real-world experiments on the task of planar pushing, we show that our method is more efficient and achieves a higher manipulation accuracy than previous vision-based approaches. |
first_indexed | 2024-09-23T15:04:18Z |
format | Article |
id | mit-1721.1/138353.2 |
institution | Massachusetts Institute of Technology |
language | English |
last_indexed | 2024-09-23T15:04:18Z |
publishDate | 2021 |
publisher | Institute of Electrical and Electronics Engineers (IEEE) |
record_format | dspace |
spelling | mit-1721.1/138353.22021-12-07T18:30:18Z Accurate Vision-based Manipulation through Contact Reasoning Bauza, Maria Wu, Jiajun Tenenbaum, Joshua B Rodriguez, Alberto Massachusetts Institute of Technology. Department of Brain and Cognitive Sciences. © 2020 IEEE. Planning contact interactions is one of the core challenges of many robotic tasks. Optimizing contact locations while taking dynamics into account is computationally costly and, in environments that are only partially observable, executing contact-based tasks often suffers from low accuracy. We present an approach that addresses these two challenges for the problem of vision-based manipulation. First, we propose to disentangle contact from motion optimization. Thereby, we improve planning efficiency by focusing computation on promising contact locations. Second, we use a hybrid approach for perception and state estimation that combines neural networks with a physically meaningful state representation. In simulation and real-world experiments on the task of planar pushing, we show that our method is more efficient and achieves a higher manipulation accuracy than previous vision-based approaches. Max Planck Society Toyota Research Institute United States. Office of Naval Research. Multidisciplinary University Research Initiative (N00014-16-1-2007) 2021-12-07T18:30:17Z 2021-12-07T15:57:25Z 2021-12-07T18:30:17Z 2020-04 2021-12-07T15:29:26Z Article http://purl.org/eprint/type/JournalArticle 1050-4729 https://hdl.handle.net/1721.1/138353.2 Kloss, Alina, Bauza, Maria, Wu, Jiajun, Tenenbaum, Joshua B, Rodriguez, Alberto et al. 2020. "Accurate Vision-based Manipulation through Contact Reasoning." Proceedings - IEEE International Conference on Robotics and Automation. en 10.1109/ICRA40945.2020.9197409 Proceedings - IEEE International Conference on Robotics and Automation Creative Commons Attribution-Noncommercial-Share Alike http://creativecommons.org/licenses/by-nc-sa/4.0/ application/octet-stream Institute of Electrical and Electronics Engineers (IEEE) arXiv |
spellingShingle | Bauza, Maria Wu, Jiajun Tenenbaum, Joshua B Rodriguez, Alberto Accurate Vision-based Manipulation through Contact Reasoning |
title | Accurate Vision-based Manipulation through Contact Reasoning |
title_full | Accurate Vision-based Manipulation through Contact Reasoning |
title_fullStr | Accurate Vision-based Manipulation through Contact Reasoning |
title_full_unstemmed | Accurate Vision-based Manipulation through Contact Reasoning |
title_short | Accurate Vision-based Manipulation through Contact Reasoning |
title_sort | accurate vision based manipulation through contact reasoning |
url | https://hdl.handle.net/1721.1/138353.2 |
work_keys_str_mv | AT bauzamaria accuratevisionbasedmanipulationthroughcontactreasoning AT wujiajun accuratevisionbasedmanipulationthroughcontactreasoning AT tenenbaumjoshuab accuratevisionbasedmanipulationthroughcontactreasoning AT rodriguezalberto accuratevisionbasedmanipulationthroughcontactreasoning |