We consider the problem of multi-agent navigation and collision avoidance when observations are limited to the local neighborhood of each agent. We propose InforMARL, a novel architecture for multi-agent reinforcement learning (MARL) which uses local information intelligently to compute paths for all the agents in a decentralized manner. Specifically, InforMARL aggregates information about the local neighborhood of agents for both the actor and the critic using a graph neural network and can be used in conjunction with any standard MARL algorithm. We show that (1) in training, InforMARL has better sample efficiency and performance than baseline approaches, despite using less information, and (2) in testing, it scales well to environments with arbitrary numbers of agents and obstacles. We illustrate these results using four task environments, including one with predetermined goals for each agent, and one in which the agents collectively try to cover all goals. Code available at https://github.com/nsidn98/InforMARL.

翻译：我们考虑在观测局限于每个智能体局部邻域时的多智能体导航与避碰问题。我们提出InforMARL，一种新颖的多智能体强化学习（MARL）架构，该架构通过智能利用局部信息以去中心化方式计算所有智能体的路径。具体而言，InforMARL利用图神经网络为演员（actor）和评论家（critic）聚合智能体局部邻域信息，并可结合任意标准MARL算法使用。我们证明：（1）在训练阶段，尽管使用较少信息，InforMARL在样本效率与性能上均优于基线方法；（2）在测试阶段，它能够良好地扩展到包含任意数量智能体和障碍物的环境。我们通过四个任务环境验证了这些结果，包括每个智能体具有预定目标的环境，以及智能体共同尝试覆盖所有目标的环境。代码见https://github.com/nsidn98/InforMARL。