RoboDream: 用于可扩展机器人数据合成的组合世界模型
RoboDream: Compositional World Models for Scalable Robot Data Synthesis
AI总结：
提出一种以具身为中心的组合世界模型，通过将轨迹执行与环境合成解耦，实现从新视角、新场景和新物体中合成逼真演示数据，并展示其在数据扩展和减少真实数据需求方面的有效性。
摘要：
扩展机器人学习需要大规模、多样化的演示，然而通过远程操作收集真实世界数据仍然过于昂贵和耗时。虽然视频扩散模型为数据扩展提供了一条有希望的途径，但现有的生成方法通常局限于表面的视觉增强，或者遭受产生物理不可行运动的具身幻觉。我们提出了一种可泛化的以具身为中心的世界模型，通过合成具有新物体、新场景和新视角的逼真演示来实现可扩展的数据生成。我们的方法将生成锚定到渲染的机器人运动，同时以显式的场景和物体先验为条件，有效地将轨迹执行与环境合成解耦。这种公式有可能解锁两种强大的数据扩展能力：（1）检索与重生，将现有轨迹重新用于全新的上下文而无需新的运动数据；（2）无道具远程操作，操作员操纵空空气，模型随后幻觉出目标物体和场景，消除了重置时间。我们通过真实世界实验证明，我们生成的数据持续改进下游策略性能，并在各种操作任务中显著减少真实世界数据需求。
英文摘要：
Scaling robot learning requires large-scale, diverse demonstrations, yet real-world data collection via teleoperation remains prohibitively expensive and time-consuming. While video diffusion models offer a promising avenue for data scaling, existing generative approaches are often limited to superficial visual augmentation, or suffer from embodiment hallucinations that yield physically infeasible motions. We present a generalizable embodiment-centric world model that achieves scalable data generation by synthesizing photorealistic demonstrations with novel objects, in novel scenes, and from novel viewpoints. Our approach anchors generation to rendered robot motion while conditioning on explicit scene and object priors, effectively decoupling trajectory execution from environment synthesis. This formulation has the potential to unlock two powerful data scaling capabilities: (1) retrieval and rebirth, which repurposes existing trajectories into entirely new contexts without new motion data; and (2) prop-free teleoperation, where operators manipulate empty air and the model hallucinates the target objects and scene afterwards, eliminating reset time. We demonstrate with real-world experiments that our generated data consistently improves downstream policy performance and significantly reduces real-world data requirements across diverse manipulation tasks.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
