Realistic modeling of agents in crowd simulations

M.W. Baig, E.I. Barakova, C.S. Regazzoni, G.W.M. Rauterberg

Research output: Chapter in Book/Report/Conference proceedingConference contributionAcademicpeer-review

7 Citations (Scopus)
2 Downloads (Pure)

Abstract

Simulation of crowd behaviors is a widely discussed topic for over a decade for computer games, architecture and entertainment. The most important thing is that agents should execute the human like behaviors in real life scenarios. This situation becomes very complex with dense crowds. Our proposed Realistic Behavior Agent Simulation (RBAS) model simulates the agents from the knowledge of empirical findings of crowd physical behaviors and crowd psychology. This paper presents a novel technique to design the agents to with following modules 1) Path planning behavior for collision avoidance 2) Situation awareness during herding behavior and turbulent flow in high density crowds. 3) Personal Reaction bubble (PRB) based response and perceptions. The evaluation with real life situations is performed to validate the RBAS model. The RBAS model allows the agents to use cognitive understandings to plan ahead their path using the visual perception information. This model encapsulates the emergent crowd behaviors such as self-organization behavior in herding situation. During turbulent high density flows, most of the existing models fail to predict the behavior of agents, the evaluations show the RBAS model mimics the same behavior of the crowd in different situations.
Original languageEnglish
Title of host publicationProceedings 5th International Conference on Intelligent Systems, Modelling and Simulation (ISMS2014), 27-29 Januari 2014, Langkawi, Malaysia
Place of PublicationPiscataway
PublisherInstitute of Electrical and Electronics Engineers
Pages507-512
ISBN (Electronic)978-1-4799-3858-2
DOIs
Publication statusPublished - 2014
Event5th International Conference on Intelligent Systems, Modelling and Simulation (ISMS2014), 27-29 Januari 2014, Langkawi, Malaysia - Langkawi, Malaysia
Duration: 27 Jan 201429 Jan 2014

Conference

Conference5th International Conference on Intelligent Systems, Modelling and Simulation (ISMS2014), 27-29 Januari 2014, Langkawi, Malaysia
CountryMalaysia
CityLangkawi
Period27/01/1429/01/14

Fingerprint

Computer games
Collision avoidance
Motion planning
Turbulent flow

Cite this

Baig, M. W., Barakova, E. I., Regazzoni, C. S., & Rauterberg, G. W. M. (2014). Realistic modeling of agents in crowd simulations. In Proceedings 5th International Conference on Intelligent Systems, Modelling and Simulation (ISMS2014), 27-29 Januari 2014, Langkawi, Malaysia (pp. 507-512). Piscataway: Institute of Electrical and Electronics Engineers. https://doi.org/10.1109/ISMS.2014.93
Baig, M.W. ; Barakova, E.I. ; Regazzoni, C.S. ; Rauterberg, G.W.M. / Realistic modeling of agents in crowd simulations. Proceedings 5th International Conference on Intelligent Systems, Modelling and Simulation (ISMS2014), 27-29 Januari 2014, Langkawi, Malaysia. Piscataway : Institute of Electrical and Electronics Engineers, 2014. pp. 507-512
@inproceedings{4fa9f16e188447fc89a19432e8f75fe6,
title = "Realistic modeling of agents in crowd simulations",
abstract = "Simulation of crowd behaviors is a widely discussed topic for over a decade for computer games, architecture and entertainment. The most important thing is that agents should execute the human like behaviors in real life scenarios. This situation becomes very complex with dense crowds. Our proposed Realistic Behavior Agent Simulation (RBAS) model simulates the agents from the knowledge of empirical findings of crowd physical behaviors and crowd psychology. This paper presents a novel technique to design the agents to with following modules 1) Path planning behavior for collision avoidance 2) Situation awareness during herding behavior and turbulent flow in high density crowds. 3) Personal Reaction bubble (PRB) based response and perceptions. The evaluation with real life situations is performed to validate the RBAS model. The RBAS model allows the agents to use cognitive understandings to plan ahead their path using the visual perception information. This model encapsulates the emergent crowd behaviors such as self-organization behavior in herding situation. During turbulent high density flows, most of the existing models fail to predict the behavior of agents, the evaluations show the RBAS model mimics the same behavior of the crowd in different situations.",
author = "M.W. Baig and E.I. Barakova and C.S. Regazzoni and G.W.M. Rauterberg",
year = "2014",
doi = "10.1109/ISMS.2014.93",
language = "English",
pages = "507--512",
booktitle = "Proceedings 5th International Conference on Intelligent Systems, Modelling and Simulation (ISMS2014), 27-29 Januari 2014, Langkawi, Malaysia",
publisher = "Institute of Electrical and Electronics Engineers",
address = "United States",

}

Baig, MW, Barakova, EI, Regazzoni, CS & Rauterberg, GWM 2014, Realistic modeling of agents in crowd simulations. in Proceedings 5th International Conference on Intelligent Systems, Modelling and Simulation (ISMS2014), 27-29 Januari 2014, Langkawi, Malaysia. Institute of Electrical and Electronics Engineers, Piscataway, pp. 507-512, 5th International Conference on Intelligent Systems, Modelling and Simulation (ISMS2014), 27-29 Januari 2014, Langkawi, Malaysia, Langkawi, Malaysia, 27/01/14. https://doi.org/10.1109/ISMS.2014.93

Realistic modeling of agents in crowd simulations. / Baig, M.W.; Barakova, E.I.; Regazzoni, C.S.; Rauterberg, G.W.M.

Proceedings 5th International Conference on Intelligent Systems, Modelling and Simulation (ISMS2014), 27-29 Januari 2014, Langkawi, Malaysia. Piscataway : Institute of Electrical and Electronics Engineers, 2014. p. 507-512.

Research output: Chapter in Book/Report/Conference proceedingConference contributionAcademicpeer-review

TY - GEN

T1 - Realistic modeling of agents in crowd simulations

AU - Baig, M.W.

AU - Barakova, E.I.

AU - Regazzoni, C.S.

AU - Rauterberg, G.W.M.

PY - 2014

Y1 - 2014

N2 - Simulation of crowd behaviors is a widely discussed topic for over a decade for computer games, architecture and entertainment. The most important thing is that agents should execute the human like behaviors in real life scenarios. This situation becomes very complex with dense crowds. Our proposed Realistic Behavior Agent Simulation (RBAS) model simulates the agents from the knowledge of empirical findings of crowd physical behaviors and crowd psychology. This paper presents a novel technique to design the agents to with following modules 1) Path planning behavior for collision avoidance 2) Situation awareness during herding behavior and turbulent flow in high density crowds. 3) Personal Reaction bubble (PRB) based response and perceptions. The evaluation with real life situations is performed to validate the RBAS model. The RBAS model allows the agents to use cognitive understandings to plan ahead their path using the visual perception information. This model encapsulates the emergent crowd behaviors such as self-organization behavior in herding situation. During turbulent high density flows, most of the existing models fail to predict the behavior of agents, the evaluations show the RBAS model mimics the same behavior of the crowd in different situations.

AB - Simulation of crowd behaviors is a widely discussed topic for over a decade for computer games, architecture and entertainment. The most important thing is that agents should execute the human like behaviors in real life scenarios. This situation becomes very complex with dense crowds. Our proposed Realistic Behavior Agent Simulation (RBAS) model simulates the agents from the knowledge of empirical findings of crowd physical behaviors and crowd psychology. This paper presents a novel technique to design the agents to with following modules 1) Path planning behavior for collision avoidance 2) Situation awareness during herding behavior and turbulent flow in high density crowds. 3) Personal Reaction bubble (PRB) based response and perceptions. The evaluation with real life situations is performed to validate the RBAS model. The RBAS model allows the agents to use cognitive understandings to plan ahead their path using the visual perception information. This model encapsulates the emergent crowd behaviors such as self-organization behavior in herding situation. During turbulent high density flows, most of the existing models fail to predict the behavior of agents, the evaluations show the RBAS model mimics the same behavior of the crowd in different situations.

U2 - 10.1109/ISMS.2014.93

DO - 10.1109/ISMS.2014.93

M3 - Conference contribution

SP - 507

EP - 512

BT - Proceedings 5th International Conference on Intelligent Systems, Modelling and Simulation (ISMS2014), 27-29 Januari 2014, Langkawi, Malaysia

PB - Institute of Electrical and Electronics Engineers

CY - Piscataway

ER -

Baig MW, Barakova EI, Regazzoni CS, Rauterberg GWM. Realistic modeling of agents in crowd simulations. In Proceedings 5th International Conference on Intelligent Systems, Modelling and Simulation (ISMS2014), 27-29 Januari 2014, Langkawi, Malaysia. Piscataway: Institute of Electrical and Electronics Engineers. 2014. p. 507-512 https://doi.org/10.1109/ISMS.2014.93