Dynamic Graph-Search Algorithm for Global Path Planning in Presence of Hazardous Weather

This paper discusses several alternatives, based on graph search, to calculate UAV trajectories that avoid regions with dangerous weather effects and with the least deviation from the shortest trajectory. It is also explained how to adapt graph search algorithms to be used for 3D trajectories with UAVs, and which design considerations should be taken into account (for example the discretization of the airspace). Also, an extension of the Lazy Theta* algorithm is presented; a dynamic algorithm that calculates the trajectory while the weather hazards information is updated. The paper includes simulations and experimental results in the CATEC testbed with multiple UAVs.