Learning Manipulation by Predicting Interaction
Fuente:
arXiv
Gespeichert in:
| Hauptverfasser: | , , , , , , , , , , , , , , |
|---|---|
| Format: | Preprint |
| Veröffentlicht: |
2024
|
| Schlagworte: | |
| Online-Zugang: | |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| _version_ | 1866911898704805888 |
|---|---|
| author | Zeng, Jia Bu, Qingwen Wang, Bangjun Xia, Wenke Chen, Li Dong, Hao Song, Haoming Wang, Dong Hu, Di Luo, Ping Cui, Heming Zhao, Bin Li, Xuelong Qiao, Yu Li, Hongyang |
| author_facet | Zeng, Jia Bu, Qingwen Wang, Bangjun Xia, Wenke Chen, Li Dong, Hao Song, Haoming Wang, Dong Hu, Di Luo, Ping Cui, Heming Zhao, Bin Li, Xuelong Qiao, Yu Li, Hongyang |
| contents | Representation learning approaches for robotic manipulation have boomed in recent years. Due to the scarcity of in-domain robot data, prevailing methodologies tend to leverage large-scale human video datasets to extract generalizable features for visuomotor policy learning. Despite the progress achieved, prior endeavors disregard the interactive dynamics that capture behavior patterns and physical interaction during the manipulation process, resulting in an inadequate understanding of the relationship between objects and the environment. To this end, we propose a general pre-training pipeline that learns Manipulation by Predicting the Interaction (MPI) and enhances the visual representation.Given a pair of keyframes representing the initial and final states, along with language instructions, our algorithm predicts the transition frame and detects the interaction object, respectively. These two learning objectives achieve superior comprehension towards "how-to-interact" and "where-to-interact". We conduct a comprehensive evaluation of several challenging robotic tasks.The experimental results demonstrate that MPI exhibits remarkable improvement by 10% to 64% compared with previous state-of-the-art in real-world robot platforms as well as simulation environments. Code and checkpoints are publicly shared at https://github.com/OpenDriveLab/MPI. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2406_00439 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | Learning Manipulation by Predicting Interaction Zeng, Jia Bu, Qingwen Wang, Bangjun Xia, Wenke Chen, Li Dong, Hao Song, Haoming Wang, Dong Hu, Di Luo, Ping Cui, Heming Zhao, Bin Li, Xuelong Qiao, Yu Li, Hongyang Robotics Computer Vision and Pattern Recognition Representation learning approaches for robotic manipulation have boomed in recent years. Due to the scarcity of in-domain robot data, prevailing methodologies tend to leverage large-scale human video datasets to extract generalizable features for visuomotor policy learning. Despite the progress achieved, prior endeavors disregard the interactive dynamics that capture behavior patterns and physical interaction during the manipulation process, resulting in an inadequate understanding of the relationship between objects and the environment. To this end, we propose a general pre-training pipeline that learns Manipulation by Predicting the Interaction (MPI) and enhances the visual representation.Given a pair of keyframes representing the initial and final states, along with language instructions, our algorithm predicts the transition frame and detects the interaction object, respectively. These two learning objectives achieve superior comprehension towards "how-to-interact" and "where-to-interact". We conduct a comprehensive evaluation of several challenging robotic tasks.The experimental results demonstrate that MPI exhibits remarkable improvement by 10% to 64% compared with previous state-of-the-art in real-world robot platforms as well as simulation environments. Code and checkpoints are publicly shared at https://github.com/OpenDriveLab/MPI. |
| title | Learning Manipulation by Predicting Interaction |
| topic | Robotics Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2406.00439 |