外观
论文arXiv 2608.302371 个月前
Motus2：用于灵巧操作的自进化通用世界模型
原文链接：Motus2: A Self-Evolving General World Model for Dexterous Manipulation（arXiv 2608.30237 · 发布日期：2026-08-31）
Motus2: A Self-Evolving General World Model for Dexterous Manipulation
Motus2面向灵巧操作，将策略、模拟器和评估器耦合为闭环，并在仿生平台上实例化。
核心要点
- 现有世界模型多在模拟器上附加动作输出头，缺少用于策略改进的闭环决策与学习耦合。
- Motus2用共享权重的单一模型提供策略、动作条件世界模型和价值模型三种控制接口。
- 数据扩展从大规模单目自我中心数据推进到同步双目数据，并结合机器人轨迹和人机对齐数据适配。
- 系统研究滑动窗口上下文扩展与触觉反馈，并在具备双目、双臂、双灵巧手和触觉的仿生平台上实例化。
验证范围与局限：摘要仅说明在具备双目视觉、双臂、双灵巧手和触觉感知的仿生平台上实例化，未给出具体任务或定量验证结果。
英文摘要 (Abstract)
General embodied agents should perceive, predict, act, evaluate, and improve within a unified system. World models have shown great promise in building such agents, yet existing models typically append an action output head to a world simulator, without coupling them into a closed decision-and-learning loop for policy improvement. We present Motus2, a self-evolving general world model for dexterous manipulation. Motus2 advances world modeling through model scaling and data scaling. For model scaling, a single model with shared weights exposes three control interfaces: a policy (world-action model), a simulator (action-conditioned world model), and an evaluator (value model). The policy proposes candidate action chunks, the simulator predicts their visual consequences, and the evaluator assesses the predicted outcomes. Their coupling forms a closed decision-and-learning loop for policy improvement. This formulation uses curated expert demonstrations for action learning, while failed and suboptimal interactions provide valuable evidence for dynamics modeling and value learning. For data scaling, Motus2 progresses from large-scale monocular egocentric data to synchronized stereo egocentric data, followed by robot-domain adaptation with robot trajectories and supplementary human-robot alignment data. Motus2 further studies global-autoregressive and hybrid-memory extensions of its sliding-window context, adds tactile feedback for contact-aware control, and is instantiated on a fully biomimetic platform with stereo vision, dual arms, dual dexterous hands, and tactile sensing. Together, egocentric data scaling and closed-loop general world model scaling provide a general path toward self-evolving dexterous manipulation.
作者：Hongzhe Bi, Zihao Zhou, Yihang Tang, Jin’an Pang, Shuhe Huang, Haitian Liu, Runqing Wang, Shuai Huang, Yichen Wang, Yiming Cheng
来源
本页事实对应的公开页面，以原文为准。