Training visual control policies from scratch on a new robot typically requires generating large amounts of robot-specific data. How might we leverage data previously collected on another robot to reduce or even completely remove this need for robot-specific data? We propose a "robot-aware control" paradigm that achieves this by exploiting readily available knowledge about the robot. We then instantiate this in a robot-aware model-based RL policy by training modular dynamics models that couple a transferable, robot-aware world dynamics module with a robot-specific, potentially analytical, robot dynamics module. This also enables us to set up visual planning costs that separately consider the robot agent and the world. Our experiments on tabletop manipulation tasks with simulated and real robots demonstrate that these plug-in improvements dramatically boost the transferability of visual model-based RL policies, even permitting zero-shot transfer of visual manipulation skills onto new robots. Project website: https://www.seas.upenn.edu/~hued/rac
翻译:从零到新机器人的培训视觉控制政策通常要求生成大量机器人特定数据。 我们如何利用先前在另一个机器人上收集的数据来减少甚至完全消除对机器人特定数据的需求? 我们提出一个“ 机器人- 观测控制” 模式,通过利用关于机器人的现成知识来实现这一点。 然后,我们通过培训模块动态模型,将一个可转让的、机器人- 认识的世界动态模块与一个机器人特定、可能进行分析的机器人动态模块结合起来,在机器人特定数据中进行同步化。这也使我们能够设置视觉规划成本,分别考虑机器人代理人和世界。 我们在桌面上与模拟和真实机器人的实验显示,这些插件操作任务大大增强了基于视觉模型的RL政策的可转让性,甚至允许将视觉操作技能零发转让给新的机器人。 项目网站: https://www.seas.upenn.edu/~hued/rac。