RT-Affordance: Affordances are Versatile Intermediate Representations for Robot Manipulation
Fuente:
arXiv
Saved in:
| Main Authors: | , , , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2024
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866909377767669760 |
|---|---|
| author | Nasiriany, Soroush Kirmani, Sean Ding, Tianli Smith, Laura Zhu, Yuke Driess, Danny Sadigh, Dorsa Xiao, Ted |
| author_facet | Nasiriany, Soroush Kirmani, Sean Ding, Tianli Smith, Laura Zhu, Yuke Driess, Danny Sadigh, Dorsa Xiao, Ted |
| contents | We explore how intermediate policy representations can facilitate generalization by providing guidance on how to perform manipulation tasks. Existing representations such as language, goal images, and trajectory sketches have been shown to be helpful, but these representations either do not provide enough context or provide over-specified context that yields less robust policies. We propose conditioning policies on affordances, which capture the pose of the robot at key stages of the task. Affordances offer expressive yet lightweight abstractions, are easy for users to specify, and facilitate efficient learning by transferring knowledge from large internet datasets. Our method, RT-Affordance, is a hierarchical model that first proposes an affordance plan given the task language, and then conditions the policy on this affordance plan to perform manipulation. Our model can flexibly bridge heterogeneous sources of supervision including large web datasets and robot trajectories. We additionally train our model on cheap-to-collect in-domain affordance images, allowing us to learn new tasks without collecting any additional costly robot trajectories. We show on a diverse set of novel tasks how RT-Affordance exceeds the performance of existing methods by over 50%, and we empirically demonstrate that affordances are robust to novel settings. Videos available at https://snasiriany.me/rt-affordance |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2411_02704 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | RT-Affordance: Affordances are Versatile Intermediate Representations for Robot Manipulation Nasiriany, Soroush Kirmani, Sean Ding, Tianli Smith, Laura Zhu, Yuke Driess, Danny Sadigh, Dorsa Xiao, Ted Robotics Artificial Intelligence Computation and Language Computer Vision and Pattern Recognition Machine Learning We explore how intermediate policy representations can facilitate generalization by providing guidance on how to perform manipulation tasks. Existing representations such as language, goal images, and trajectory sketches have been shown to be helpful, but these representations either do not provide enough context or provide over-specified context that yields less robust policies. We propose conditioning policies on affordances, which capture the pose of the robot at key stages of the task. Affordances offer expressive yet lightweight abstractions, are easy for users to specify, and facilitate efficient learning by transferring knowledge from large internet datasets. Our method, RT-Affordance, is a hierarchical model that first proposes an affordance plan given the task language, and then conditions the policy on this affordance plan to perform manipulation. Our model can flexibly bridge heterogeneous sources of supervision including large web datasets and robot trajectories. We additionally train our model on cheap-to-collect in-domain affordance images, allowing us to learn new tasks without collecting any additional costly robot trajectories. We show on a diverse set of novel tasks how RT-Affordance exceeds the performance of existing methods by over 50%, and we empirically demonstrate that affordances are robust to novel settings. Videos available at https://snasiriany.me/rt-affordance |
| title | RT-Affordance: Affordances are Versatile Intermediate Representations for Robot Manipulation |
| topic | Robotics Artificial Intelligence Computation and Language Computer Vision and Pattern Recognition Machine Learning |
| url | https://arxiv.org/abs/2411.02704 |