Path planning for efficient UAV coordination in aerobiological sampling missions

This paper is concerned with the coordinated flight of two autonomous UAVs to be used for aerobiological sampling of biological threat agents above agricultural fields. The periodic sampling task involves two phases: a sampling interval and an initialization interval. During the sampling interval, both vehicles must employ their aerobiological sampling devices and follow a precise ground track in the presence of sustained winds. During the initialization interval, the vehicles move to their respective initial states to begin the next sampling interval. To maximize the volume of air sampled by the UAVs during an individual sampling mission, the initialization interval must be as short as possible. The paper provides a simple, geometric method for generating candidate time optimal paths in steady winds, based on Dubins¿ well-known results for minimum time paths of bounded curvature. The approach is used to generate paths for both UAVs, which must coordinate their motion along their respective paths in order to avoid collision. The described methods were tested during an aerobiological sampling experiment focusing on the plant pathogen Phytophthora infestans.