Multi-Agent Navigation in Dynamic Environments

We study the problem of coordinating teams of vehicles with limited sensing and communication to navigate in environments with dynamic (adversarial) and static obstacles using graph neural networks and multi-agent reinforcement learning.


  1. InforMARL: Scalable Multi-Agent Reinforcement Learning through Intelligent Information Aggregation.


Siddharth Nayak, Sydney Dolan, Jasmine Jerry Aloor, Allen Shtofenmakher, Victor Qin, Geoffrey Ding, Kenneth Choi, Wenqi Ding, Hamsa Balakrishnan