RoboClaw: An Agentic Framework for Scalable Long-Horizon Robotic Tasks
Fuente:
arXiv
Saved in:
| Main Authors: | , , , , , , , , , , , , , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2026
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866911559317454848 |
|---|---|
| author | Li, Ruiying Zhou, Yunlang Zhu, YuYao Chen, Kylin Wang, Jingyuan Wang, Sukai Hu, Kongtao Yu, Minhui Jiang, Bowen Su, Zhan Ma, Jiayao He, Xin Shen, Yongjian Yang, Yang Ren, Guanghui Yao, Maoqing Wang, Wenhao Mu, Yao |
| author_facet | Li, Ruiying Zhou, Yunlang Zhu, YuYao Chen, Kylin Wang, Jingyuan Wang, Sukai Hu, Kongtao Yu, Minhui Jiang, Bowen Su, Zhan Ma, Jiayao He, Xin Shen, Yongjian Yang, Yang Ren, Guanghui Yao, Maoqing Wang, Wenhao Mu, Yao |
| contents | Vision-Language-Action (VLA) systems have shown strong potential for language-driven robotic manipulation. However, scaling them to long-horizon tasks remains challenging. Existing pipelines typically separate data collection, policy learning, and deployment, resulting in heavy reliance on manual environment resets and brittle multi-policy execution. We present RoboClaw, an agentic robotics framework that unifies data collection, policy learning, and task execution under a single VLM-driven controller. At the policy level, RoboClaw introduces Entangled Action Pairs (EAP), which couple forward manipulation behaviors with inverse recovery actions to form self-resetting loops for autonomous data collection. This mechanism enables continuous on-policy data acquisition and iterative policy refinement with minimal human intervention. During deployment, the same agent performs high-level reasoning and dynamically orchestrates learned policy primitives to accomplish long-horizon tasks. By maintaining consistent contextual semantics across collection and execution, RoboClaw reduces mismatch between the two phases and improves multi-policy robustness. Experiments in real-world manipulation tasks demonstrate improved stability and scalability compared to conventional open-loop pipelines, while significantly reducing human effort throughout the robot lifecycle, achieving a 25% improvement in success rate over baseline methods on long-horizon tasks and reducing human time investment by 53.7%. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2603_11558 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | RoboClaw: An Agentic Framework for Scalable Long-Horizon Robotic Tasks Li, Ruiying Zhou, Yunlang Zhu, YuYao Chen, Kylin Wang, Jingyuan Wang, Sukai Hu, Kongtao Yu, Minhui Jiang, Bowen Su, Zhan Ma, Jiayao He, Xin Shen, Yongjian Yang, Yang Ren, Guanghui Yao, Maoqing Wang, Wenhao Mu, Yao Robotics Artificial Intelligence Vision-Language-Action (VLA) systems have shown strong potential for language-driven robotic manipulation. However, scaling them to long-horizon tasks remains challenging. Existing pipelines typically separate data collection, policy learning, and deployment, resulting in heavy reliance on manual environment resets and brittle multi-policy execution. We present RoboClaw, an agentic robotics framework that unifies data collection, policy learning, and task execution under a single VLM-driven controller. At the policy level, RoboClaw introduces Entangled Action Pairs (EAP), which couple forward manipulation behaviors with inverse recovery actions to form self-resetting loops for autonomous data collection. This mechanism enables continuous on-policy data acquisition and iterative policy refinement with minimal human intervention. During deployment, the same agent performs high-level reasoning and dynamically orchestrates learned policy primitives to accomplish long-horizon tasks. By maintaining consistent contextual semantics across collection and execution, RoboClaw reduces mismatch between the two phases and improves multi-policy robustness. Experiments in real-world manipulation tasks demonstrate improved stability and scalability compared to conventional open-loop pipelines, while significantly reducing human effort throughout the robot lifecycle, achieving a 25% improvement in success rate over baseline methods on long-horizon tasks and reducing human time investment by 53.7%. |
| title | RoboClaw: An Agentic Framework for Scalable Long-Horizon Robotic Tasks |
| topic | Robotics Artificial Intelligence |
| url | https://arxiv.org/abs/2603.11558 |