让机器人在复杂动作中稳定应对异常指令，避免崩溃。
CMP: Robust Whole-Body Tracking for Loco-Manipulation via Competence Manifold Projection
- 通过能力流形投影，将安全约束转化为单步计算的几何限制。
- 在异常指令下生存率提升10倍，轨迹偏差小于10%。
- 可自动适应未知任务，实现渐进式尽力完成目标。
尽管分步控制策略在足式移动操作机上表现出鲁棒性，但学习整体全身控制策略以跟踪末端执行器全局位姿，在传感器噪声或不可行用户指令引发的分布外（OOD）输入下仍易失效。为在不牺牲任务性能和连续性的前提下增强对扰动的鲁棒性，我们提出能力流形投影（CMP）。具体地，采用逐帧安全机制，将无限时域安全约束转化为计算高效的单步流形包含。为构建该能力流形，引入下界安全估计算法，区分未掌握意图与训练分布。进一步设计同构隐空间（ILS），使流形几何与安全概率对齐，实现O(1)时间复杂度的无缝防御。实验表明，CMP在典型OOD场景中生存率提升10倍，轨迹误差低于10%。系统还展现出自发的‘尽力而为’泛化行为，逐步完成分布外目标。视频演示见：https://shepherd1226.github.io/CMP。
原文摘要 · Abstract (English)
While decoupled control schemes for legged mobile manipulators have shown robustness, learning holistic whole-body control policies for tracking global end-effector poses remains fragile against Out-of-Distribution (OOD) inputs induced by sensor noise or infeasible user commands. To improve robustness against these perturbations without sacrificing task performance and continuity, we propose Competence Manifold Projection (CMP). Specifically, we utilize a Frame-Wise Safety Scheme that transforms the infinite-horizon safety constraint into a computationally efficient single-step manifold inclusion. To instantiate this competence manifold, we employ a Lower-Bounded Safety Estimator that distinguishes unmastered intentions from the training distribution. We then introduce an Isomorphic Latent Space (ILS) that aligns manifold geometry with safety probability, enabling efficient O(1) seamless defense against arbitrary OOD intents. Experiments demonstrate that CMP achieves up to a 10-fold survival rate improvement in typical OOD scenarios where baselines suffer catastrophic failure, incurring under 10% tracking degradation. Notably, the system exhibits emergent ``best-effort'' generalization behaviors to progressively accomplish OOD goals by adhering to the competence boundaries. Result videos are available at: https://shepherd1226.github.io/CMP.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。