论文标题
Bususbot：学习在繁忙的板环境中学习互动，推理和计划
BusyBot: Learning to Interact, Reason, and Plan in a BusyBoard Environment
论文作者
论文摘要
我们介绍了忙碌的板，这是一种受玩具启发的机器人学习环境，它利用一组各种铰接的对象和对象间功能关系为机器人交互提供丰富的视觉反馈。基于这种环境，我们介绍了一个学习框架，即Bushusbot，该框架允许代理商以综合和自我监督的方式共同获得三个基本功能（互动，推理和计划）。借助Busy Board提供的丰富感官反馈，Busybot首先学习了有效与环境互动的政策；然后，随着使用该策略收集的数据，Busybot的原因是通过因果发现网络对象间功能关系；最后，通过结合学习的互动政策和关系推理技能，代理可以执行目标条件的操纵任务。我们在模拟环境和现实环境中评估了忙碌的机器人，并验证了其看不见的对象和关系的普遍性。视频可从https://youtu.be/ej98xbjz9ek获得。
We introduce BusyBoard, a toy-inspired robot learning environment that leverages a diverse set of articulated objects and inter-object functional relations to provide rich visual feedback for robot interactions. Based on this environment, we introduce a learning framework, BusyBot, which allows an agent to jointly acquire three fundamental capabilities (interaction, reasoning, and planning) in an integrated and self-supervised manner. With the rich sensory feedback provided by BusyBoard, BusyBot first learns a policy to efficiently interact with the environment; then with data collected using the policy, BusyBot reasons the inter-object functional relations through a causal discovery network; and finally by combining the learned interaction policy and relation reasoning skill, the agent is able to perform goal-conditioned manipulation tasks. We evaluate BusyBot in both simulated and real-world environments, and validate its generalizability to unseen objects and relations. Video is available at https://youtu.be/EJ98xBJZ9ek.
