Decentralized Multi-agent Path Finding in Dynamic Environments

In many potential real-world applications of mobile robot fleets, such as e-commerce warehouses (Amazon), the robots must be able to operate efficiently in a dynamic environment where obstacles stochastically appear. While the related Multi-Agent Pathfinding (MAPF) problem is widely studied, existing planners mainly rely on offline planning as well as simplistic assumptions that make them ill-suited for real-life dynamic environments. 

We expand the application domain for efficient warehouse robots to non-highly controlled environments. For this purpose, we propose a decentralized approach that can coordinate large fleets of mobile robots to operate effectively in dynamic environments.