Emerging Inter-Swarm Collaboration for Surveillance Using Pheromones and Evolutionary Techniques

• Published in: Sensors (Basel, Switzerland) Vol. 20; no. 9; p. 2566
• Main Authors: Stolfi, Daniel H, Brust, Matthias R, Danoy, Grégoire, Bouvry, Pascal
• Format: Journal Article
• Language: English
• Published: Switzerland MDPI AG 30-04-2020; MDPI
• Subjects: Algorithms; Autonomous vehicles; Collaboration; Communication; Control algorithms; Cooperation; evolutionary algorithm; Evolutionary algorithms; Genetic algorithms; Health surveillance; inter-swarm collaboration; mobility model; Pheromones; Robotics; Robots; swarm robotics; unmanned aerial vehicle; Unmanned aerial vehicles; unmanned ground vehicle; Unmanned ground vehicles; Wireless networks; inter-swarm collaboration; unmanned aerial vehicle; bio-inspiration; mobility model; pheromones; swarm robotics; unmanned ground vehicle; evolutionary algorithm
• ISSN: 1424-8220; 1424-8220
• DOI: 10.3390/s20092566

In this article, we propose a new mobility model, called Attractor Based Inter-Swarm collaborationS (ABISS), for improving the surveillance of restricted areas performed by unmanned autonomous vehicles. This approach uses different types of vehicles which explore an area of interest following unpredictable trajectories based on chaotic solutions of dynamic systems. Collaborations between vehicles are meant to cover some regions of the area which are unreachable by members of one swarm, e.g., unmanned ground vehicles on water surface, by using members of another swarm, e.g., unmanned aerial vehicles. Experimental results demonstrate that collaboration is not only possible but also emerges as part of the configurations calculated by a specially designed and parameterised evolutionary algorithm. Experiments were conducted on 12 different case studies including 30 scenarios each, observing an improvement in the total covered area up to 11%, when comparing ABISS with a non-collaborative approach.