Skip to content

论文HF Daily Papers · arXiv 2611.1706010 天前

Ropedia Xperience 2发布:1589个3D扫描与4230条第一人称数据构建具身S。 ​

本文由 AI 根据公开内容整理,可能有误,请以原文为准:Ropedia Xperience 2: A Gold Standard for Scaling Physical AI Environments and Behaviors(HF Daily Papers · arXiv 2611.17060)

核心内容 ​

尽管GPT-6 Astra等大模型已能通过视觉直接控制机器人,但在真实世界长程接触任务中仍表现脆弱。为解决物理AI能力如何随环境与行为规模扩展的问题,研究团队推出Ropedia Xperience 2数据集。该资产包含1589个度量级3D环境扫描和4230条第一人称人类演示视频,覆盖74种任务类型,总时长达131小时,为建立具身智能Scaling Law提供了标准化的数据基础。

技术上,该数据集实现了环境采集与行为数据的联合复用:3D扫描被转化为具备物理交互属性的仿真环境,人类运动通过重定向生成全身机器人轨迹。此外,系统支持通过虚拟遥操作和自动化长程轨迹生成获取额外机器人行为数据。这种将真实场景数字化并同步映射到机器人策略空间的流程,验证了从人类演示到机器人仿真训练的闭环可行性,解决了高质量交互环境稀缺的工程瓶颈。

关键看点 ​

  • 数据规模:开源1589个度量级3D扫描与4230条第一人称视频,覆盖74类任务共131小时,提供标准化物理AI训练资产
  • 仿真转换:将真实3D扫描自动转化为物理交互式仿真环境,支持策略在闭环中部署、评估与迭代优化
  • 动作迁移:利用运动重定向技术将人类第一人称演示映射为全身机器人轨迹,结合虚拟遥操作生成长程行为数据
  • Scaling Law:提出环境与行为联合采集范式,旨在解决物理AI缺乏类似语言模型明确扩展规律的核心难题

工程与产业观察 ​

该工作确立了“环境+行为”联合采集作为具身数据金标准,通过高保真3D扫描与动作重定向打通Real-to-Sim链路。对于算法工程师,这提供了验证Scaling Law的高质量基准;对产业界而言,其标准化的数据生产管线降低了构建大规模交互仿真环境的边际成本,加速了长程操作策略的泛化研究。

英文摘要 (Abstract) ​

Large foundation models such as GPT-6 Astra and Claude Fable 5.1 have recently been shown to control robots directly from visual observations. They nevertheless remain brittle on long-horizon, contact-rich tasks in real-world environments. Further progress depends on two complementary resources: demonstrations of human behavior that supervise how tasks are performed, and interactive environments in which policies can be rolled out, evaluated, and improved in closed loop. Language models have advanced along well-characterized scaling laws; the corresponding question for physical AI is how capability scales with environments and behaviors, which first requires identifying which data to scale and how to produce and reuse such data continuously. To address this question, we introduce Ropedia Xperience 2, comprising 1,589 released metric-scale 3D scans and a collection of 4,230 egocentric recordings of 74 tasks (131 hours). We further show that each capture can be repurposed for robot learning: scenes are converted into physically interactive simulation environments, human motion is retargeted to whole-body robot trajectories, and additional robot behaviors are obtained through virtual teleoperation and automated long-horizon trajectory generation. We propose this joint capture of environments and behaviors as a gold standard for physical AI data collection, laying a data foundation for establishing scaling laws over environments and behaviors.

Hugging Face Daily Papers 热度:18 upvotes 作者:Hao Li, Fangzhou Hong, Zhaoxi Chen, Hui En Pang, Kairui Hu, Mingze Sun, Runmao Yao, Xinhao Ji, Xiyuan Zheng, Xuanli Chen, Yalun Dai, Yap Kim Hui, Dingwen Zhang, Yuanyuan Gao, Ziang Cao, Cheng Huang Ye…

基于本页技术方向关联的整机参数、核心零部件、实测报告与同赛道选型横评。

外部来源

本页事实对应的公开页面,以原文为准。(点击将离开本站)

  1. HF Daily Papers · arXiv 2611.17060:Ropedia Xperience 2: A Gold Standard for Scaling Physical AI Environments and Behaviors ↗访问 2026-11-13