Social interactions in crowds of pedestrians: An adaptive model for group cohesion

Abstract

The paper introduces an agent-based model for the simulation of crowds of pedestrians whose main innovative aspect is the representation and management of an important type of social interaction among the pedestrians: members of groups, in fact, carry out of a form of interaction (by means of verbal or non-verbal communication) that allows them to preserve the cohesion of the group even in particular conditions, such as counter flows, presence of obstacles or narrow passages. The paper formally describes the model and presents both qualitative and quantitative results in sample simulation scenarios. © Springer International Publishing Switzerland 2013.