A pack hunting strategy for heterogeneous robots in rescue operations.

This study focuses on improving coordination among teams of heterogeneous robots, including Unmanned Aerial Vehicles (UAVs) and Unmanned Ground Vehicles (UGVs), drawing inspiration from natural pack-hunting strategies. The goal is to increase the effectiveness of rescue operations using a new framework that combines hierarchical decision making with decentralised control. The approach features dynamic target assignment and real time task allocation based on a scoring function that considers multiple factors, such as the distance to the target, energy usage, communication ability, and potential for energy exchange. In contrast to methods that use static roles, this system allows robots to change between 'Chaser' and 'Flanker' roles based on current data, improving adaptability. Results showed that this approach led to better coordination and decision-making, with robots autonomously adjusting their roles to improve mission outcomes. The findings suggest that combining hierarchical structures with decentralised control improves responsiveness and ensures the effective use of resources in complex, changing environments, making this method well-suited for real-world rescue operations.