ZeroMimic：从网络视频蒸馏机器人操作技能

近年来，机器人操作学习的许多进展都来自于模仿学习，但这些方法大多依赖于来自同一机器人、同一房间、且在测试时必须处理相同对象和环境的演示。这类演示难以获取。相比之下，已存在大量在野外展示操作技能的人类视频数据集，包含了对机器人有价值的信息。我们是否可以没有任何机器人专用演示或探索要求，从这些数据中提炼出一套有用的机器人技能策略？我们提出了首个能够实现此目标的系统ZeroMimic，它能够为几类常见的操作任务（打开、关闭、倒酒、抓取与放置、切割、搅拌）生成可立即部署的图像目标条件技能策略，这些策略能够处理多样化的对象并适应各种未遇到的任务环境。ZeroMimic carefully designed to exploit recent advances in semantic and geometric visual understanding of human videos, together with modern grasp affordance detectors and imitation policy classes. After training ZeroMimic on the popular EpicKitchens dataset of ego-centric human videos, we evaluate its out-of-the-box performance in varied real-world and simulated kitchen settings with two different robot embodiments, demonstrating its impressive abilities to handle these varied tasks. To enable plug-and-play reuse of ZeroMimic policies on other task setups and robots, we release software and policy checkpoints of our skill policies.