Bots of a feather: Exploring user perceptions of group cohesiveness for application in robotic swarms

Stower, Rebecca, Zibetti, Elisabetta et St-Onge, David. 2022. « Bots of a feather: Exploring user perceptions of group cohesiveness for application in robotic swarms ». In 31st IEEE International Symposium on Robot and Human Interactive Communication (RO-MAN) (Naples, Italie, Aug. 29 - Sept. 02, 2022) pp. 95-100. IEEE.

Prévisualisation

PDF St-Onge-D-2022-25168.pdf - Version acceptée Licence d'utilisation : Tous les droits réservés aux détenteurs du droit d'auteur. Télécharger (495kB) | Prévisualisation

Résumé

Behaviours of robot swarms often take inspiration from biological models, such as ant colonies and bee hives. Yet, understanding how these behaviours are actually perceived by human users has so far received limited attention. In this paper, we use animations to represent different kinds of possible swarm motions intended to communicate specific messages to a human. We explore how these animations relate to the perceived group cohesiveness of the swarm, comprised of five different parameters: synchronising, grouping, following, reacting, and shape forming. We conducted an online user study where 98 participants viewed nine animations of a swarm displaying different behaviours and rated them for perceived group cohesiveness. We found that the parameters of group cohesiveness correlated with the messages the swarm was perceived as communicating. In particular, the message of initiating communication was highly positively correlated with all group parameters, whereas broken communication was negatively correlated. In addition, the importance of specific group parameters differed within each animation. For example, the parameter of grouping was most associated with animations signalling an intervention is needed. These findings are discussed within the context of designing intuitive behaviour for robot swarms.

Type de document:	Compte rendu de conférence
Professeur:	Professeur St-Onge, David
Affiliation:	Génie mécanique
Date de dépôt:	12 sept. 2022 14:30
Dernière modification:	12 déc. 2022 21:05
URI:	https://espace2.etsmtl.ca/id/eprint/25168

Actions (Authentification requise)

Dernière vérification avant le dépôt