Discretization and OpenCL-based implementation of social field pedestrian model

Yu, Bin; Wu, Kuanghuai; Zhu, Ke; Ji, Yuhan; Zhang, H. Michael; Xia, Yang

doi:10.1016/j.neucom.2018.06.074

Cited by 5 publications

(15 citation statements)

References 22 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Except allowing the continuous model to own a larger degree of freedom, using the concept of field also leads to a straightforward discretization. As shown in the work of [24], a discrete model was derived. Especially, the concept of walk period was introduced.…”

Section: Continuous and Discrete Modelsmentioning

confidence: 99%

“…To map the social field model into the OpenCL heterogeneous framework, factors including those already discussed in the work of [24] are considered so far. Thus it is meaningful to have a summary (table III), among which the issue of competence is more to say.…”

Section: Bank Conflictmentioning

confidence: 99%

“…In order to examine the current GPU model's numerical efficiency, the scenarios used in the work of [24] are re-experimented. Firstly, the discrete space runs from 100 × 100, 200 × 200, · · · , to 1000 × 1000.…”

Section: Numerical Experimentsmentioning

confidence: 99%

“…In addition to the better modeling of crowd dynamics, the proposed continuous model has the advantage of easy discretization so that a discrete version and a later OpenCL-based implementation were developed. Yu et al [24] reported that this can bring an at most 30.8 times speedup with comparison to the CPU model.…”

Section: Literature Reviewmentioning

confidence: 99%

“…The paper is the follow-up research and tries to improve the work of [24] in two aspects. Firstly, it develops a method based on the idea of divide-andconquer to solve the problem of global memory depletion when fields have a large geometric size.…”

Section: Literature Reviewmentioning

confidence: 99%

See 4 more Smart Citations

Improved OpenCL-Based Implementation of Social Field Pedestrian Model

Zhu

et al. 2020

IEEE Trans. Intell. Transport. Syst.

Self Cite

View full text Add to dashboard Cite

Two aspects of improvements are proposed for the OpenCL-based implementation of the social field pedestrian model. In the aspect of algorithm, a method based on the idea of divide-and-conquer is devised in order to overcome the problem of global memory depletion when fields are of a larger size. This is of importance for the study of finer pedestrian walking behavior, which usually requires larger fields. In the aspect of computation, the OpenCL heterogeneous framework is thoroughly studied. Factors that may affect the numerical efficiency are evaluated, with regarding to the social field model previously proposed. This includes usage of local memory, deliberate patch of data structures for avoidance of bank conflicts, and so on. Numerical experiments disclose that the numerical efficiency is brought to an even higher level. Compared to the CPU model and the previous GPU model, the current GPU model can be at most 71.56 and 13.3 times faster respectively so that it is more qualified to be a core engine for analysis of super-large scale crowd.

show abstract

Section: Continuous and Discrete Modelsmentioning

confidence: 99%

Section: Bank Conflictmentioning

confidence: 99%

Section: Numerical Experimentsmentioning

confidence: 99%

Section: Literature Reviewmentioning

confidence: 99%

Section: Literature Reviewmentioning

confidence: 99%

See 3 more Smart Citations

Improved OpenCL-Based Implementation of Social Field Pedestrian Model

Zhu

et al. 2020

IEEE Trans. Intell. Transport. Syst.

Self Cite

View full text Add to dashboard Cite

show abstract

Modeling of crowd groups with an extended social field model

Yu¹

2023

J. Stat. Mech.

Self Cite

View full text Add to dashboard Cite

A social field model is extended by adding supports of simulations of crowd group related dynamics such as group cohesion, back-tracking, etc. A computational implementation combining heterogeneous computing and multi-threading technologies is developed to integrate the proposed extension into a heterogeneous computing framework. Hence modeling of group related dynamics can be accomplished in a very efficient manner. In the section of numerical experiments, the extended model is firstly validated with a set of practical data. Results show that the simulated evacuation time matches the practical one quite well. Then a study of fundamental diagrams is expanded with considerations of crowd groups. It is shown that the impact of crowd groups mainly happens in the regime of low densities and would become insignificant in the regime of high densities.

show abstract

Integrated modeling of crowd groups based on spatial distribution of members

Yu,

Ou,

2024

J. Stat. Mech.

Self Cite

View full text Add to dashboard Cite

An integrated framework supporting both leader-follower and leaderless groups is proposed. The concept of subgroup is supported so that chains of groups can exist. This allows to model complicated group relationships in crowd simulations. To model group related dynamics, rules are proposed. A detailed explanation of implementation of these rules is given. To validate the proposed framework, two sets of numerical experiments are conducted, using experimental data collected by previous researchers. In one set of experiments, small leader-follower groups without subgroups are considered. In the other set, groups are leaderless and can be small or large with zero or more subgroups. The impact of crowd groups on fundamental diagrams is also experimented. The conducted experiments indicate that the proposed framework can reproduce results that are quite comparable to practical values, thus the framework’s efficacy is validated within a certain range.

show abstract

Discretization and OpenCL-based implementation of social field pedestrian model

Cited by 5 publications

References 22 publications

Improved OpenCL-Based Implementation of Social Field Pedestrian Model

Improved OpenCL-Based Implementation of Social Field Pedestrian Model

Modeling of crowd groups with an extended social field model

Integrated modeling of crowd groups based on spatial distribution of members

Contact Info

Product

Resources

About