arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2610.36013cs.CLcs.CV
收藏

OmniBrain-0:一种用于具身智能的统一智能体模型

OmniBrain-0: A Unified Agentic Model for Embodied Intelligence

University of Macau(澳门大学);Tsinghua University(清华大学)
Xiaomi Embodied Intelligence Team, Yuechen Luo, Chong He, Fang Li, Shaoqing Xu, Jingkai Wang, Menglun Su, Jiaxin Liu, Shengyin Jiang, Zhiyi Lai, Long Chen, Long… 展开作者

Xiaomi Embodied Intelligence Team, Yuechen Luo, Chong He, Fang Li, Shaoqing Xu, Jingkai Wang, Menglun Su, Jiaxin Liu, Shengyin Jiang, Zhiyi Lai, Long Chen, Longlong Wang, Liang Hu, Haochen Liu, Yuncheng Jiang, Hongwei Xie, Jieming Cui, Hanbing Li, Junjie Huang, Zhixiang Duan, Hexin Zhang, Weixiang Liang, Ying Gong, Yuhan Wang, Yong Pan, Pengna Li, Kangyi Wu, Ken Deng, Kun Ma, Fuxi Wen, Zhi-xin Yang, Hangjun Ye

AI总结:

研究针对通用具身智能的长时程技能协调与故障恢复难题,提出轻量级统一智能体OmniBrain-0,结合渐进式训练框架与RSI机制,在多基准及真实任务上实现高性能且低开销的具身能力整合。

摘要:

通用具身智能要求机器人在长时程中协调多种技能、跟踪任务进度并从故障中恢复。仅靠专用策略无法保证任务级可靠性,而强大的多模态智能体依赖高昂的推理成本和外部经验,紧凑模型难以复现这些特性。我们提出OmniBrain-0,一种轻量级统一具身智能体,它通过共享的高层决策模型协调操作与导航,整合了任务理解、规划、进度跟踪及反馈驱动的恢复功能。我们的渐进式训练框架结合了具身预训练、领域专用智能体蒸馏以及多教师同策略蒸馏(MOPD),该框架将强大教师的交互经验蒸馏为操作和导航专家,再将其决策能力整合至OmniBrain-0。因此,统一发生在学习到的决策模型内部,而非仅在工具接口处。我们还提出了基于规则的递归自我改进(RSI),这是一种数据飞轮机制,其中元智能体将执行证据转化为优化后的规则、步骤级提醒以及经验证的轨迹,用于领域专家更新,之后这些专家再通过MOPD重新整合至OmniBrain-0。我们在LIBERO-Pro、LIBERO-Plus、RoboCasa365、RoboTwin 2.0、R2R-CE Val-Unseen以及多个真实世界操作任务上对OmniBrain-0进行了评估。OmniBrain-0兼具强大的跨域任务性能,且推理和执行开销远低于其闭源教师,证明了具身能力可高效迁移并整合至紧凑的统一智能体中。

英文摘要:

General-purpose embodied intelligence requires robots to coordinate diverse skills over long horizons, track task progress, and recover from failures. Specialist policies alone do not ensure task-level reliability, while powerful multimodal agents rely on costly reasoning and external experience that compact models struggle to reproduce. We introduce OmniBrain-0, a lightweight unified embodied agent that coordinates manipulation and navigation through a shared high-level decision model, integrating task understanding, planning, progress tracking, and feedback-driven recovery. Our progressive training framework combines embodied pretraining, domain-specialized agent distillation, and multi-teacher on-policy distillation (MOPD). It distills powerful teachers' interaction experience into manipulation and navigation experts, then consolidates their decision capabilities into OmniBrain-0. Unification therefore occurs within the learned decision model rather than only at the tool interface. We further introduce Rubric-Guided Recursive Self-Improvement (RSI), a data flywheel in which a Meta-Agent turns execution evidence into refined rubrics, step-level reminders, and verified trajectories for domain-expert updates before their reconsolidation into OmniBrain-0 through MOPD. We evaluate OmniBrain-0 on LIBERO-Pro, LIBERO-Plus, RoboCasa365, RoboTwin 2.0, and R2R-CE Val-Unseen, as well as multiple real-world manipulation tasks. OmniBrain-0 combines strong cross-domain task performance with substantially lower inference and execution overhead than its closed-source teacher, demonstrating efficient transfer and consolidation of embodied capabilities into a compact unified agent.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑