Robotic manipulation in complex open-world scenarios requires both reliable physical manipulation skills and effective and generalizable perception. In this paper, we propose a method where general purpose pretrained visual models serve as an object-centric prior for the perception system of a learned policy. We devise an object-level attentional mechanism that can be used to determine relevant objects from a few demonstrations, and then immediately incorporate those objects into a learned policy. A task-independent meta-attention locates possible objects in the scene, and a task-specific attention identifies which objects are predictive of the demonstrations. The scope of the task-specific attention is easily adjusted by showing demonstrations with distractor objects or with diverse relevant objects. Our results indicate that this approach exhibits good generalization across object instances using very few samples, and can be used to learn a variety of manipulation tasks using reinforcement learning.

本文提出了一种方法来解决复杂开放环境下机器人操作的问题，该方法基于先前训练的通用视觉模型作为感知系统的对象先验，并引入了一个基于对象的注意机制来确定相关对象，通过少数轨迹或演示将这些对象纳入学习策略，使用强化学习可以学习多种操作任务。

面向通用机器人学习的深度目标中心表示