Self-organised Aggregation in Swarms of Robots with Informed Robots

Ziya FIRAT, Eliseo Ferrante, Nicholas Cambier, Elio Tuci

Résultats de recherche: Contribution dans un livre/un catalogue/un rapport/dans les actes d'une conférenceArticle dans les actes d'une conférence/un colloque

1 Téléchargements (Pure)

Résumé

In this paper, we study a swarm of robots that has to select one aggregation site in an environment in which two sites are available. It is known in the literature that, in presence of asymmetries in the environment, robot swarms are able to perform a collective choice and aggregate in one among two possible sites, for example the largest of the two. We focus on an aggregation scenario where the environment is morphologically symmetric. The two aggregation sites are identical with only one exception: their colour. In addition, in the swarm only a proportion of robots, that we call the informed robots, possess extra information concerning on which specific site the swarm is required to aggregate. The rest of the robots are non-informed, thus they do not possess the above mentioned extra information. In simulation-based experiments we show that, if no robot in the swarm is informed, the swarm is able to break the symmetry and aggregates on one of the two sites at random. However, the introduction of a small proportion of informed robots is enough to break the symmetry: the majority of the swarm aggregates on the site preferred by the informed robot. Additionally, the swarm is also able to completely aggregate on one of the two sites when only 30% of the robots are informed, independently from the swarm size among those we considered. Finally, we analyse how the time dynamics of the aggregation process depend on the proportion of informed robots.

langue originaleAnglais
titreTheory and Practice of Natural Computing
Sous-titre7th International Conference, TPNC 2018, Proceedings
rédacteurs en chefCarlos Martín-Vide, Miguel A. Vega-Rodríguez, David Fagan, Michael O’Neill
EditeurSpringer Verlag
Pages49-60
Nombre de pages12
ISBN (Electronique)978-3-030-04070-3
ISBN (imprimé)978-3-030-04069-7
Les DOIs
Etat de la publicationPublié - 2018

Série de publications

NomLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume11324 LNCS
ISSN (imprimé)0302-9743
ISSN (Electronique)1611-3349

Empreinte digitale

Examiner les sujets de recherche de « Self-organised Aggregation in Swarms of Robots with Informed Robots ». Ensemble, ils forment une empreinte digitale unique.

Contient cette citation