Multi-agent search strategy based on digital pheromones for UAVs

Abstract

Unmanned Aerial Vehicles (UAVs) are steadily replacing conventional manned aerial vehicles in multitude of tasks. Many of those tasks can be performed more efficiently by using a coordination between a team of UAVs rather than a single UAV. In this paper problem of cooperative area coverage using team of UAVs is addressed. Digital pheromones are used to convey information about search area. UAVs communicate indirectly by depositing and sensing digital pheromones which exist in pheromone map. Two approaches for path planning using pheromone maps are proposed. First approach is based on genetic algorithm, whereas second one relies on roulette wheel selection. Both approaches are evaluated in terms of distance required to cover given percentage of the area.