GR-RL: Going Dexterous and Precise for Long-Horizon Robotic Manipulation
Fuente:
arXiv
Enregistré dans:
| Auteurs principaux: | , , , , , , , , , , , , , , , , , , , , |
|---|---|
| Format: | Preprint |
| Publié: |
2025
|
| Sujets: | |
| Accès en ligne: | |
| Tags: |
Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
|
| _version_ | 1866914216579956736 |
|---|---|
| author | Li, Yunfei Ma, Xiao Xu, Jiafeng Cui, Yu Cui, Zhongren Han, Zhigang Huang, Liqun Kong, Tao Liu, Yuxiao Niu, Hao Peng, Wanli Qiao, Jingchao Ren, Zeyu Shi, Haixin Su, Zhi Tian, Jiawen Xiao, Yuyang Zhang, Shenyu Zheng, Liwei Li, Hang Wu, Yonghui |
| author_facet | Li, Yunfei Ma, Xiao Xu, Jiafeng Cui, Yu Cui, Zhongren Han, Zhigang Huang, Liqun Kong, Tao Liu, Yuxiao Niu, Hao Peng, Wanli Qiao, Jingchao Ren, Zeyu Shi, Haixin Su, Zhi Tian, Jiawen Xiao, Yuyang Zhang, Shenyu Zheng, Liwei Li, Hang Wu, Yonghui |
| contents | We present GR-RL, a robotic learning framework that turns a generalist vision-language-action (VLA) policy into a highly capable specialist for long-horizon dexterous manipulation. Assuming the optimality of human demonstrations is core to existing VLA policies. However, we claim that in highly dexterous and precise manipulation tasks, human demonstrations are noisy and suboptimal. GR-RL proposes a multi-stage training pipeline that filters, augments, and reinforces the demonstrations by reinforcement learning. First, GR-RL learns a vision-language-conditioned task progress, filters the demonstration trajectories, and only keeps the transitions that contribute positively to the progress. Specifically, we show that by directly applying offline RL with sparse reward, the resulting $Q$-values can be treated as a robust progress function. Next, we introduce morphological symmetry augmentation that greatly improves the generalization and performance of GR-RL. Lastly, to better align the VLA policy with its deployment behaviors for high-precision control, we perform online RL by learning a latent space noise predictor. With this pipeline, GR-RL is, to our knowledge, the first learning-based policy that can autonomously lace up a shoe by threading shoelaces through multiple eyelets with an 83.3% success rate, a task requiring long-horizon reasoning, millimeter-level precision, and compliant soft-body interaction. We hope GR-RL provides a step toward enabling generalist robot foundation models to specialize into reliable real-world experts. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2512_01801 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | GR-RL: Going Dexterous and Precise for Long-Horizon Robotic Manipulation Li, Yunfei Ma, Xiao Xu, Jiafeng Cui, Yu Cui, Zhongren Han, Zhigang Huang, Liqun Kong, Tao Liu, Yuxiao Niu, Hao Peng, Wanli Qiao, Jingchao Ren, Zeyu Shi, Haixin Su, Zhi Tian, Jiawen Xiao, Yuyang Zhang, Shenyu Zheng, Liwei Li, Hang Wu, Yonghui Robotics Machine Learning We present GR-RL, a robotic learning framework that turns a generalist vision-language-action (VLA) policy into a highly capable specialist for long-horizon dexterous manipulation. Assuming the optimality of human demonstrations is core to existing VLA policies. However, we claim that in highly dexterous and precise manipulation tasks, human demonstrations are noisy and suboptimal. GR-RL proposes a multi-stage training pipeline that filters, augments, and reinforces the demonstrations by reinforcement learning. First, GR-RL learns a vision-language-conditioned task progress, filters the demonstration trajectories, and only keeps the transitions that contribute positively to the progress. Specifically, we show that by directly applying offline RL with sparse reward, the resulting $Q$-values can be treated as a robust progress function. Next, we introduce morphological symmetry augmentation that greatly improves the generalization and performance of GR-RL. Lastly, to better align the VLA policy with its deployment behaviors for high-precision control, we perform online RL by learning a latent space noise predictor. With this pipeline, GR-RL is, to our knowledge, the first learning-based policy that can autonomously lace up a shoe by threading shoelaces through multiple eyelets with an 83.3% success rate, a task requiring long-horizon reasoning, millimeter-level precision, and compliant soft-body interaction. We hope GR-RL provides a step toward enabling generalist robot foundation models to specialize into reliable real-world experts. |
| title | GR-RL: Going Dexterous and Precise for Long-Horizon Robotic Manipulation |
| topic | Robotics Machine Learning |
| url | https://arxiv.org/abs/2512.01801 |