Lifelong Multi Agent Path Finding
The project explores the Multi-Agent Pathfinding (MAPF) problem, focusing on one-shot MAPF and its dynamic counterpart, Lifelong MAPF (LMAPF). It delves into the Conflict-Based Search Algorithm (CBS), a two-level approach for resolving conflicts between agents, showcasing its effectiveness in small-scale environments inspired by the League of Robot Runners competition.
Brief Explanation of the Project (Read full Report here):
Approach:
The approach centers around the Conflict-Based Search Algorithm (CBS), a novel method for resolving agent conflicts. CBS operates on two levels: a high level that forms a search tree for conflicts and a low level that finds paths for individual agents. The report compares CBS with traditional pathfinding algorithms, demonstrating its efficiency and scalability in dynamic environments.
Results:
Experimental results showcase CBS's performance in two different environments: a random map and a warehouse map. These tests reveal the algorithm's strength in handling complex multi-agent scenarios with minimal conflicts and delays. The warehouse map, simulating a real-world logistics scenario, particularly highlights CBS's practical applicability in managing dynamic agent paths efficiently.
Read Full Report here