Agent collaboration and social networks

Bowman, R. Sean; Hexmoor, Henry

doi:10.1109/kimas.2005.1427082

Cited by 3 publications

(1 citation statement)

References 1 publication

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Despite potential benefits of a social network‐based approach to agent coordination, there has been only limited research in social aspects of embodied agents, such as the pioneering work of Dautenhahn (1995), Dautenhahn and Christaller (1996), Ogden and Dautenhahn (2001) and recently Bowman and Hexmoor (2005).…”

Section: Social Network In Multi‐robot Systemsmentioning

confidence: 99%

Social networks in simulated multi‐robot environment

Rodic

Engelbrecht

2008

International Journal of Intelligent Computing and Cybernetics

View full text Add to dashboard Cite

PurposeThe purpose of this paper is to present a novel approach to coordination of multi‐agent teams, and in particular multi‐robot teams. The new approach is based on models of organisational sociology, namely the concept of social networks. The social relationships used in the model that is presented in this paper are trust and kinship relationships, but modified for use in heterogeneous multi‐robot teams.Design/methodology/approachThe coordination of a robot team is achieved through task allocation. The proposed task allocation mechanism was tested in the multi‐robot team task allocation simulation.FindingsThe social networks‐based task allocation algorithm has performed according to expectations and the obtained results are very promising. Some intriguing similarities with higher mammalian societies were observed and they are discussed in this paper. The social networks‐based approach also exhibited the ability to learn and store information using social networks.Research limitations/implicationsThe research focused on simulated environments and further research is envisaged in the physical environments to confirm the applicability of the presented approach.Practical implicationsIn this paper, the proposed coordination was applied to simulated multi‐robot teams. It is important to note that the proposed coordination model is not robot specific, but can also be applied to almost any multi‐agent system without major modifications.Originality/valueThe paper emphasizes applicability of considering multi‐robot teams as socially embodied agents. It also presents a novel and efficient approach to task allocation.

show abstract

Section: Social Network In Multi‐robot Systemsmentioning

confidence: 99%