Sim2Real Manipulation on Unknown Objects with Tactile-based Reinforcement Learning
Fuente:
arXiv
Guardado en:
| Autores principales: | , , , , , , |
|---|---|
| Formato: | Preprint |
| Publicado: |
2024
|
| Materias: | |
| Acceso en línea: | |
| Etiquetas: |
Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
|
| _version_ | 1866910371595419648 |
|---|---|
| author | Su, Entong Jia, Chengzhe Qin, Yuzhe Zhou, Wenxuan Macaluso, Annabella Huang, Binghao Wang, Xiaolong |
| author_facet | Su, Entong Jia, Chengzhe Qin, Yuzhe Zhou, Wenxuan Macaluso, Annabella Huang, Binghao Wang, Xiaolong |
| contents | Using tactile sensors for manipulation remains one of the most challenging problems in robotics. At the heart of these challenges is generalization: How can we train a tactile-based policy that can manipulate unseen and diverse objects? In this paper, we propose to perform Reinforcement Learning with only visual tactile sensing inputs on diverse objects in a physical simulator. By training with diverse objects in simulation, it enables the policy to generalize to unseen objects. However, leveraging simulation introduces the Sim2Real transfer problem. To mitigate this problem, we study different tactile representations and evaluate how each affects real-robot manipulation results after transfer. We conduct our experiments on diverse real-world objects and show significant improvements over baselines for the pivoting task. Our project page is available at https://tactilerl.github.io/. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2403_12170 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | Sim2Real Manipulation on Unknown Objects with Tactile-based Reinforcement Learning Su, Entong Jia, Chengzhe Qin, Yuzhe Zhou, Wenxuan Macaluso, Annabella Huang, Binghao Wang, Xiaolong Robotics Using tactile sensors for manipulation remains one of the most challenging problems in robotics. At the heart of these challenges is generalization: How can we train a tactile-based policy that can manipulate unseen and diverse objects? In this paper, we propose to perform Reinforcement Learning with only visual tactile sensing inputs on diverse objects in a physical simulator. By training with diverse objects in simulation, it enables the policy to generalize to unseen objects. However, leveraging simulation introduces the Sim2Real transfer problem. To mitigate this problem, we study different tactile representations and evaluate how each affects real-robot manipulation results after transfer. We conduct our experiments on diverse real-world objects and show significant improvements over baselines for the pivoting task. Our project page is available at https://tactilerl.github.io/. |
| title | Sim2Real Manipulation on Unknown Objects with Tactile-based Reinforcement Learning |
| topic | Robotics |
| url | https://arxiv.org/abs/2403.12170 |