Multi-Agents Path Planning for a Swarm of Unmanned Aerial Vehicles

Abstract

The objective of swarms is to enable multiple agents to collaborate toward a common goal, as one would find in a remote sensing setting. In this paper we focus on swarms of unmanned aerial vehicles (UAVs), which for instance have an objective to optimize the survey of a prescribed area and/or the detection of a specific object. Instructing each individual agent from a central command control quickly becomes inefficient, even for small groups of agents. Agreement protocol is done locally by the multi-agents without external user input. Because of the wide variety of conditions UAVs can face, the algorithm needs to be robust despite external disturbances.