Applying Conflict-Based Search to Improve Pathfinding Efficiency in Multi-Agent Big Data Adventure Games

The implementation of intelligent pathfinding algorithms is essential for improving gameplay experience and educational value in educational games. This study looks into applying conflict-based search for multi-agent pathfinding in an educational game called "Data Quest: Big Data Adventure," intended for active learning in teaching big data concepts. Conflict-based Search (CBS) optimizes pathfinding and conflict resolution through multi-agent coordination, thereby improving real-time responsiveness and educational engagement. CBS enhances real-time response and education. The results show that compared with conventional approaches such as cooperative A*, CBS reduces the computation time of pathfinding by 30-50% in medium to high-density environments and achieves a conflict resolution efficiency of 75% to 85% in the first iteration. The generated paths by CBS are nearly optimal, differing by only 5-10% from ideal paths, which is adequate in educational scenarios. The successful application of CBS not only enhances the performance of multi-agent interactions but also underscores the importance of integrating advanced AI techniques in educational gaming, paving the way for more immersive and effective learning experiences. These results show how well CBS can enable dynamic and complex interaction within an educational game, highlighting its potential for broader educational applications that benefit from advanced AI-driven, multi-agent systems.