发表机构
Fudan University; AI Research Center, Midea Group (Shanghai) Co., Ltd.; Carnegie Mellon University（复旦大学; 美的集团（上海）有限公司AI研究中心; 卡内基梅隆大学）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
FolDeX是一个基于真实机器人数据的物理世界基准，以衣物折叠为主任务，提供2000多小时、20多任务数据，围绕数据复用四个研究轴，并建立公平评估平台，推动长时程可变形操作研究。
AI 中文摘要
具身人工智能，包括视觉-语言-动作模型和世界-动作模型，必须在物理世界中可靠地运行。然而，在仿真中表现良好的方法在真实机器人上可能会大幅退化，尤其是在长时程可变形物体操作中，策略必须跟踪不断变化的状态并执行可靠的多阶段双臂交互。现有的真实机器人基准主要集中于短时程刚体物体任务，对长时程可变形操作的覆盖有限。我们提出了FolDeX，一个完全基于真实机器人数据构建的物理世界基准，以衣物折叠为主要任务。由于真实机器人数据收集成本高昂，FolDeX研究了如何高效地复用异构物理经验。该基准围绕四个研究轴组织：利用部署期间收集的人类干预和恢复数据；跨任务迁移数据，包括跨衣物类别以及从刚体到可变形物体操作；在光照、背景和布局变化的场景中复用数据；以及跨机器人本体迁移数据。FolDeX提供了超过2000小时的真实机器人数据，涵盖20多个任务和10多种机器人本体。我们还建立了一个公平的真实机器人评估平台，用于外部提交的策略，具有标准化任务、留出物理对象、受控初始化和统一执行协议。该平台可通过此https URL公开访问。我们希望FolDeX能成为异构真实机器人数据复用和可靠长时程可变形操作统一测试平台。
英文摘要
Embodied AI, including vision-language-action and world-action models, must operate reliably in the physical world. Yet methods that perform well in simulation can degrade substantially on real robots, especially in long-horizon deformable-object manipulation, where policies must track changing states and execute reliable multi-stage bimanual interactions. Existing real-robot benchmarks mainly focus on short-horizon rigid-object tasks and offer limited coverage of long-horizon deformable manipulation. We introduce RoboFolDeX, a physical-world benchmark built entirely from real-robot data, with garment folding as its primary task. Since real-robot data collection is costly, RoboFolDeX studies how heterogeneous physical experience can be reused efficiently. The benchmark is organized around four research axes: leveraging human intervention and recovery data collected during deployment; transferring data across tasks, including across garment categories and from rigid to deformable-object manipulation; reusing data across scenes with changes in lighting, background, and layout; and transferring data across robotic embodiments. RoboFolDeX provides 2,000+ hours of real-robot data spanning 20+ tasks and 10+ embodiments. We also establish a fair real-robot evaluation platform for externally submitted policies, with standardized tasks, held-out physical objects, controlled initializations, and a unified execution protocol. The platform is publicly accessible at https://ai.midea.com/#/fold-challenge. We hope RoboFolDeX will serve as a unified testbed for heterogeneous real-robot data reuse and reliable long-horizon deformable manipulation.
