Panic Propagation Dynamics of High-Density Crowd Based on Information Entropy and Aw-Rascle Model

Autor: Daheng Dong, Rongyong Zhao, Yunlong Ma, Cuiling Li, Qianshan Hu, Qiong Liu
Rok vydání: 2020
Předmět:
Zdroj: IEEE Transactions on Intelligent Transportation Systems. 21:4425-4434
ISSN: 1558-0016
1524-9050
DOI: 10.1109/tits.2019.2953357
Popis: It is significant to discovery the impact of panic on crowd movement and study the panic propagation mechanism which can help real crowd control. This paper focuses on panic propagation dynamics in a high-density crowd based on information entropy theory and Aw-Rascle model. A novel concept of panic entropy is defined to quantify the confusion degree of crowd movement based on information entropy theory. The behavior characteristics are measured by the modulus and direction of the gridded crowd velocity vector in a two-dimension space. In order to study panic propagation dynamics, Aw-Rascle model is used to describe the crowd flow dynamics. Then, the dynamic model of panic propagation is proposed based on crowd flow. To validate the panic propagation model, numerical simulations are conducted based on one of stampedes happened in the Mecca Hajj in 2015. Simulation results show the relationship between the crowd panic entropy, density and velocity with visual panic distribution and different color contour diagrams. As a discussion result of the relationship between velocity and panic entropy, it is verified that: 1) moderate panic can help the crowd to keep relatively high velocity; 2) excessive panic can lead to irregular crowd movement and decrease crowd movement velocity especially in the over-crowded situation. The contribution of this paper is to provide a novel approach to quantify crowd panic degree and study panic propagation dynamics in a high-density crowd.
Databáze: OpenAIRE
Externí odkaz: