Adaptive hierarchical swarming in persistent surveillance applications

Persistent surveillance applications are closely related to pursuit evasion games. Global and local maximum pursuit policies are useful techniques for directing swarms of unmanned aerial vehicles (UAVs) in pursuit of ground vehicles (GVs). While global communication is often available there are costs associated with communication: range constraints, bandwidth constraints, power consumption and requirements for silent operation. A number of greedy methods for swarming are explored in the literature. Some of these methods explore potential field approaches for pheromone placement and pursuit. In this paper both potential field and discrete pheromone strategies are analyzed for the pursuit evasion problems. In addition, hierarchical techniques for directing UAVs by communications from XUAVs (UAVs with more extensive sensors) are investigated.