In this paper consensus formation in multi-agent systems is studied using the framework of kinetic theory of gases. This can be done by interpreting agents as the molecules of a gas and interactions among agents as the collisions among the molecules. Each agent can interact with any other agent in the considered system and interactions are binary, i.e., they involve only two agents at a time. Under such hypothesis, starting from a microscopic model which describes the effects of interactions in a pair of agents, we derive, analytically, global properties of the system, such as the conservation of the average opinion. Simulation results show that the proposed kinetic approach allows reaching consensus and the velocity of convergence to consensus depends, as expected, on the parameters of the model.
A study of consensus formation using kinetic theory / Monica, S., Bergenti, F. (ADVANCES IN INTELLIGENT SYSTEMS AND COMPUTING). - In: Distributed Computing and Artificial Intelligence, 13th International Conference / [a cura di] Omatu S.;Semalat A.;Bocewicz G.;Sitek P.;Nielsen I.E.;García García J.A.;Bajo J.. - GEWERBESTRASSE 11, CHAM, CH-6330, SWITZERLAND : Springer Verlag, 2016. - ISBN 9783319401614. - pp. 213-221 [10.1007/978-3-319-40162-1_23]
A study of consensus formation using kinetic theory
Monica Stefania;Bergenti Federico
2016
Abstract
In this paper consensus formation in multi-agent systems is studied using the framework of kinetic theory of gases. This can be done by interpreting agents as the molecules of a gas and interactions among agents as the collisions among the molecules. Each agent can interact with any other agent in the considered system and interactions are binary, i.e., they involve only two agents at a time. Under such hypothesis, starting from a microscopic model which describes the effects of interactions in a pair of agents, we derive, analytically, global properties of the system, such as the conservation of the average opinion. Simulation results show that the proposed kinetic approach allows reaching consensus and the velocity of convergence to consensus depends, as expected, on the parameters of the model.
Pubblicazioni consigliate
I metadati presenti in IRIS UNIMORE sono rilasciati con licenza Creative Commons CC0 1.0 Universal, mentre i file delle pubblicazioni sono rilasciati con licenza Attribuzione 4.0 Internazionale (CC BY 4.0), salvo diversa indicazione.
In caso di violazione di copyright, contattare Supporto Iris
