Chaos-enhanced mobility models for multilevel swarms of UAVs

Martin Rosalie (1) , Grégoire Danoy (2) , Serge Chaumette (3) , Pascal Bouvry (2)

1 SnT - Interdisciplinary Centre for Security, Reliability and Trust [Luxembourg]
2 CSC - Computer Science and Communications Research Unit [Luxembourg]
3 LaBRI - Laboratoire Bordelais de Recherche en Informatique

The number of civilian and military applications using Unmanned Aerial Vehicles (UAVs) has increased during the last years and the forecasts for upcoming years are exponential. One of the current major challenges consist in considering UAVs as autonomous swarms to address some limitations of single UAV usage such as autonomy, range of operation and resilience. In this article we propose novel mobility models for multi-level swarms of collaborating UAVs used for the coverage of a given area. These mobility models generate unpredictable trajectories using a chaotic solution of a dynamical system. We detail how the chaotic properties are used to structure the exploration of an unknown area and enhance the exploration part of an Ant Colony Optimization method. Empirical evidence of the improvement of the coverage efficiency obtained by our mobility models is provided via simulation. It clearly outperforms state-of-the-art approaches.

Mots clés: ACO; Swarm of UAVs; Mobility model; Chaotic dynamics; Global optimization; First return map

Domaines: Dynamique Chaotique [nlin.CD]; Systèmes dynamiques [math.DS]; Système multi-agents [cs.MA]

Martin Rosalie, Grégoire Danoy, Serge Chaumette, Pascal Bouvry. Chaos-enhanced mobility models for multilevel swarms of UAVs. Swarm and Evolutionary Computation, 2018, ⟨10.1016/j.swevo.2018.01.002⟩. ⟨hal-01699298⟩