OmniContact: 通过接触流链接元技能实现可泛化的人形移动操作
OmniContact: Chaining Meta-Skills via Contact Flow for Generalizable Humanoid Loco-Manipulation
AI总结：
提出以接触流为核心的分层框架，通过低层策略学习统一技能库、高层模块合成未来接触流序列，实现人形机器人长时程任务的鲁棒执行与自主恢复，在搬运箱子和推叠箱子任务上分别达到98.7%和76.5%的成功率。
摘要：
学习长时程人形移动操作面临双重挑战：不仅需要元技能的鲁棒执行，还需要其无缝的闭环链接并具备自主恢复能力。现有方法仍有限制：显式的人-物交互表示精度高但难以用于高层规划，而隐式技能嵌入紧凑但缺乏可靠组合所需的可解释性。我们提出\ours，一个以\textbf{接触流（CF）}为中心的分层框架，接触流是一种紧凑表示，包含关键身体轨迹和时间序列二元接触信号。利用这一共享接口，我们的低层策略\textbf{CF-Track}学习一个统一的移动操作技能库，而高层模块\textbf{CF-Gen}启发式地合成未来的接触流序列。为支持这一设置，我们还收集了OmniContact数据集，一个基于MoCap的人形移动操作HOI语料库（附录~\ref{sec:dataset}）。它们共同实现了元技能的鲁棒执行、自主故障恢复和灵活组合，以完成长时程任务。实验表明，OmniContact在\textit{搬运箱子}任务上达到\(98.7\%\)的成功率，在\textit{推叠箱子}任务上达到\(76.5\%\)的成功率，在元技能和技能链接方面分别比先前基线平均高出\(40.9\%\)和\(66.5\%\)。此外，我们的框架自然地与VLM集成，用于语义任务分解，从而实现复杂的、基于语义的移动操作行为，例如将散落的箱子排列成心形。
英文摘要：
Learning long-horizon humanoid loco-manipulation poses a dual challenge: it requires not only the robust execution of meta-skills but also their seamless, closed-loop chaining equipped with autonomous recovery. Existing approaches remain limited: explicit humanoid-object interaction representations offer precision but are notoriously difficult for high-level planning, whereas implicit skill embeddings are compact but lack the interpretability required for reliable composition. We propose \ours, a hierarchical framework centered on \textbf{contact flow (CF)}, a compact representation consisting of key body trajectories and time-series binary contact signals. Leveraging this shared interface, our low-level policy \textbf{CF-Track} learns a unified library of loco-manipulation skills, while our high-level module \textbf{CF-Gen} heuristically synthesizes future contact-flow sequences. To support this setting, we additionally collect the OmniContact dataset, a MoCap-based HOI corpus for humanoid loco-manipulation (Appendix~\ref{sec:dataset}). Together, they enable robust execution, autonomous failure recovery, and flexible composition of meta-skills for long-horizon tasks. Experiments show that OmniContact achieves \(98.7\%\) success on \textit{Carry Box} and \(76.5\%\) on \textit{Push-Stack Boxes}, outperforming prior baselines by average margins of \(40.9\%\) in meta-skill and \(66.5\%\) in skill chaining. Besides, our framework naturally integrates with VLMs for semantic task decomposition, enabling complex, semantically grounded loco-manipulation behaviors, such as arranging scattered boxes into a heart shape.
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。
