外观
论文arXiv 2610.0822022 天前
VOMMI：为移动操作收集并利用便携式示范
原文链接：VOMMI: Collecting and Leveraging Portable Demonstrations for Mobile Manipulation（arXiv 2610.08220 · 发布日期：2026-10-06）
VOMMI: Collecting and Leveraging Portable Demonstrations for Mobile Manipulation
VOMMI通过离线轨迹重建和在线视觉运动条件，将便携RGB示范用于VLA后训练，并在三项真机任务验证。
核心要点
- 便携式移动操作示范可缓解数据稀缺，但从RGB获得可靠低成本运动监督仍困难。
- VOMMI同步身体与手部视角，连接便携RGB示范与VLA后训练，无需人机运动学对应标定。
- R2-VO用稀疏几何锚点细化离线轨迹，并生成多预测时域的因果局部运动token。
- 仅用便携示范后训练时，基座速度误差低18.2%，末端平移精度与机器人示范策略相当。
验证范围与局限：实验在三项真机任务上验证；每项任务使用500条便携示范、75条留作RGB-VO评估，并以200条机器人示范作参考。
英文摘要 (Abstract)
Portable mobile-manipulation demonstrations can help alleviate data scarcity for embodied intelligence, but obtaining reliable, low-cost, and robot-free motion supervision from RGB observations remains challenging. Existing approaches often rely on teleoperation or specialized devices equipped with additional sensing hardware, while directly using estimated visual odometry (VO) trajectories can introduce inconsistencies due to accumulated drift and imperfect motion supervision. We present the Visual-Odometry-Conditioned Mobile Manipulation Interface (VOMMI), a portable demonstration collection and learning framework that connects portable RGB demonstrations to vision-language-action (VLA) post-training through offline trajectory reconstruction and online visual-motion conditioning. VOMMI synchronizes body and hand views to capture navigation context and local object interactions without requiring human-robot kinematic correspondence calibration. R2-VO refines offline demonstration trajectories using sparse geometric anchors and produces causal local-motion tokens over multiple prediction horizons for online policy conditioning. An action-group residual adapter incorporates these tokens only into the base branch. Experiments use a 500-trajectory portable for each task, with 75 trajectories held out for RGB-VO evaluation, and 200 robot demonstrations as references. Our policy, post-trained only on portable demonstrations, achieves 18.2% lower base-velocity error than a policy trained with robot-collected demonstrations, while maintaining comparable end-effector translation accuracy. Offline reconstruction reduces absolute trajectory errors for the body and hand streams by 24.6% on average relative to the best evaluated baseline for each stream. The complete system improves the mean success rate by 8.3 percentage points over OpenPI 0.5 across three real-robot tasks.
作者：Yutian Zhang, Xingrui Xiong, Siyuan Ma, Yang Li, Jiawen Wen, Jiaqi Zhai, Liwen Yang, Ce Hao, Haozhen Chi, Yangkun Zhu
来源
本页事实对应的公开页面，以原文为准。