Dynamic Task Allocation for Disaster Relief Using Multi-Agent Systems with A* and Rollout Strategy
Shuqin Zhou, Xiaofan Wang · 2025
In disaster relief scenarios, the increasing demand for rapid response and efficient resource allocation presents significant challenges. This paper proposes an intelligent material transportation algorithm based on a Multi-Agent System (MAS), integrating the classical A* pathfinding algorithm with a multi-agent rollout strategy ($R+M$) for dynamic task allocation and path planning. Experimental results demonstrate that the R+M algorithm shows significant advantages in key indicators such as total path length, number of collisions, task completion time, and success rate.