Optimized 3D mapping of a large area with structures using multiple multirotors

This paper presents an optimized method to employ a team of multirotors for collecting images to generate a 3D map of an area with emphasis on special features (e.g. man-made structures) and areas of avoidance. When an area is too large to be mapped by one multirotor within its battery life, then either several flights or several multirotors need to be employed to complete the job. This paper introduces a method for equitably dividing the task of collecting images utilizing multiple vehicles to map a large area with man-made structures, which need to be mapped with high resolution. First, the number of vehicles or flights required must be determined and then the special features within the area that require more time must be considered. A parallel flood fill algorithm incorporating game theory is used to determine the best way to divide the tasking.