In this paper, we propose an agent-based model for crowd simulation. It is made up of two types of agents that act differently: collaboratives, which share information about the paths and/or repair the ones that have been damaged, and defectors, who share no information, destroy some paths and/or nodes, and take advantage of the collaborative agents’ guidance. The aim of the model is to investigate how the agents who engage in these two activities affect one another and, ultimately, on the collective behavior of the crowd. For both kinds of agents, three evaluation metrics have been considered: (i) the number of agents that have reached the exit; (ii) paths’ costs; and (iii) exit times. According to the data analysis, the presence of defectors is essential in improving the agents’ performance. Indeed, when both collaborators and defectors are present in the crowd, more agents exit, and their average path costs and average exit times are lower than when only collaborative agents are present.

An Agent-Based Model for Crowd Simulation

Carolina Crespi;Georgia Fargetta;Mario Francesco Pavone;Rocco Alessandro Scollo
2023-01-01

Abstract

In this paper, we propose an agent-based model for crowd simulation. It is made up of two types of agents that act differently: collaboratives, which share information about the paths and/or repair the ones that have been damaged, and defectors, who share no information, destroy some paths and/or nodes, and take advantage of the collaborative agents’ guidance. The aim of the model is to investigate how the agents who engage in these two activities affect one another and, ultimately, on the collective behavior of the crowd. For both kinds of agents, three evaluation metrics have been considered: (i) the number of agents that have reached the exit; (ii) paths’ costs; and (iii) exit times. According to the data analysis, the presence of defectors is essential in improving the agents’ performance. Indeed, when both collaborators and defectors are present in the crowd, more agents exit, and their average path costs and average exit times are lower than when only collaborative agents are present.
2023
978-3-031-31182-6
978-3-031-31183-3
agent-base model; ant colony optimization; swarm intelligence; crowd simulation; collective behaviour.
File in questo prodotto:
Non ci sono file associati a questo prodotto.

I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.

Utilizza questo identificativo per citare o creare un link a questo documento: https://hdl.handle.net/20.500.11769/572736
Citazioni
  • ???jsp.display-item.citation.pmc??? ND
  • Scopus 1
  • ???jsp.display-item.citation.isi??? ND
social impact