UAV Fleet Mobility Model with Multiple Pheromones for Tracking Moving Observation Targets

• Published in: Applications of Evolutionary Computation pp. 332 - 347
• Main Authors: Atten, Christophe, Channouf, Loubna, Danoy, Grégoire, Bouvry, Pascal
• Format: Book Chapter
• Language: English
• Published: Cham Springer International Publishing 2016
• Series: Lecture Notes in Computer Science
• Subjects: Ant Colony Optimisation; Mobility model; Pheromone; Target tracking; UAV
• ISBN: 3319312030; 9783319312033
• ISSN: 0302-9743; 1611-3349
• DOI: 10.1007/978-3-319-31204-0_22

The last years, UAVs have been developed to address a variety of applications ranging from searching and tracking to the surveillance of an area. However, using a single UAV limits the range of possible applications. Therefore, fleets of UAVs are nowadays considered to work together on a common goal which requires novel distributed mobility management models. This work proposes a novel nature-inspired mobility model for UAV fleets based on Ant Colony Optimisation approaches (ACO). It relies on two types of pheromones, a repulsive pheromone to cover the designated area in an efficient way, and an attractive pheromone to detect and to track the maximum number of targets. Furthermore, all decision takings are taken online by each UAV and are fully distributed. Experimental results demonstrate promising target tracking performances together with a small increase in the exhaustivity of the coverage.