用概率流建模机器人运动速度场，实现更高效精准的物体操作。
Flow as Flow: Modeling Robot Velocity Fields as Probability Velocity Fields for Flow-Based Object Manipulation
- 将机器人运动建模为连续速度场，而非稀疏关键点位移。
- 在标准基准上生成速度场快33倍，成功率更高。
- 适合跨机器人形态的通用运动生成任务。
跨形态数据已成为训练机器人基础模型的核心。为利用这类异构数据，本文聚焦于基于流的物体操作，其中机器人流（机器人速度场）作为与具体形态无关的运动表示。以往研究将机器人流建模为稀疏关键点的位移，未能充分反映运动的连续时间特性；而本工作提出Flow as Flow框架，基于流匹配形式将机器人流建模为概率流。该方法自然地在连续速度场中进行建模，实现了高效且高质量的机器人流生成。在标准基准测试中，该方法优于代表性基线，在各项指标上表现更优，同时生成速度提升约33倍。通过在13个操作任务上对9种方法进行260次/方法的实机实验，验证了本方法平均成功率显著高于基线。项目页面见：https://flow-as-flow-u0n5y.kinsta.page。
原文摘要 · Abstract (English)
Cross-embodiment data have become central to training robotic foundation models. To leverage such heterogeneous data, we focus on flow-based object manipulation, where robot flows (robot velocity fields) serve as embodiment-agnostic motion representations. Previous studies do not formulate robot flows as dense velocity fields, but as displacements of sparse keypoints, while such velocity fields better match the continuous-time nature of motions. We propose Flow as Flow, a framework that models robot flows as probability flows based on a flow matching formulation. By naturally modeling such velocity fields within this formulation, our method achieves efficient and high-quality robot flow generation. Across standard benchmarks, our method outperforms representative baseline methods on standard metrics, while achieving approximately 33$\times$ faster generation. Furthermore, through real-world experiments evaluating 9 methods with 260 trials per method across 13 manipulation tasks, we show that our method achieves a higher average success rate than the baseline methods. Our project page is available at https://flow-as-flow-u0n5y.kinsta.page.
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。