基于高斯混合模型的受引导人群疏散研究

刘天宇; 杨晓霞; 张纪会; 赵逸群; 周美琦; 刘天宇; 杨晓霞; 张纪会; 赵逸群; 周美琦

doi:10.13306/j.1672-3813.2020.03.006

复杂系统与复杂性科学 >

2020 , Vol. 17 >Issue 3: 62 - 69

DOI: https://doi.org/10.13306/j.1672-3813.2020.03.006

基于高斯混合模型的受引导人群疏散研究

刘天宇 ,
杨晓霞 ,
张纪会 ,
赵逸群 ,
周美琦 ,
刘天宇 ,
杨晓霞 ,
张纪会 ,
赵逸群 ,
周美琦

展开

1.青岛大学自动化学院复杂性科学研究所,山东青岛 266071;
2.青岛理工大学机械与汽车工程学院,山东青岛 266525

刘天宇(1994-),男,山东烟台人,硕士研究生,主要研究方向为行人动力学模型。

收稿日期: 2020-06-01

修回日期: 2020-06-12

网络出版日期: 2020-09-23

基金资助

国家自然科学基金(61673228),山东省自然科学基金(ZR2018PF008)

收起

The Guided Crowd Evacuation Based on Gaussian Mixture Model

LIU Tianyu ,
YANG Xiaoxia ,
ZHANG Jihui ,
ZHAO Yiqun ,
ZHOU Meiqi ,
LIU Tianyu ,
YANG Xiaoxia ,
ZHANG Jihui ,
ZHAO Yiqun ,
ZHOU Meiqi

Expand

1. Institute of Complexity Science, College of Automation, Qingdao University, Qingdao 266071, China;
2. School of Mechanical and Automotive Engineering, Qingdao University of Technology, Qingdao 266525, China

Received date: 2020-06-01

Revised date: 2020-06-12

Online published: 2020-09-23

Fold

摘要

融合高斯混合模型及模糊逻辑理论,提出基于元胞自动机模型的受引导人群疏散动力学模型,用于研究引导员对疏散行为的影响。贝叶斯信息准则(BIC)确定引导员的最优数量,EM算法确定引导员的最优位置。元胞自动机模型驱动行人运动,模糊逻辑理论模拟行人对引导员的选择行为。研究了引导员的数量和速度以及出口宽度对疏散动力学的影响,并得到结论:引导员在一定程度上可以提高疏散效率,然而引导员的数量并非越多越好;一定范围内增加出口宽度能提高出口的通行能力,有效降低疏散时间;引导员速度为行人速度的75%时,疏散效率最高。本文模型融合了元胞自动机模型计算量小及高斯混合模型聚类准确度高的优势,为行人动力学建模提供新思路,对行人疏散行为提出有效的建议。

关键词： 元胞自动机模型; 高斯混合模型; 模糊逻辑理论; EM算法; 引导员

本文引用格式

刘天宇 , 杨晓霞 , 张纪会 , 赵逸群 , 周美琦 , 刘天宇 , 杨晓霞 , 张纪会 , 赵逸群 , 周美琦 . 基于高斯混合模型的受引导人群疏散研究[J]. 复杂系统与复杂性科学, 2020 , 17(3) : 62 -69 . DOI: 10.13306/j.1672-3813.2020.03.006

Abstract

In this paper, the guided crowd evacuation dynamics model based on cellular automata is proposed, this model combines Gaussian mixture methodand fuzzy logic theory to study the influence of guides on pedestrian evacuation behaviors. Bayesian Information Criterion (BIC)determines the optimal number of guides and EM algorithmdetermines the optimal positions of the guides. The cellular automata model is used as the driven model of pedestrian motion, and fuzzy logic theory is adopted to simulate the pedestrians' selection behaviors for the guides. The influences of guide quantity, guide speed and exit width on the evacuation are explored, and it is concluded that the guides can improve the evacuation efficiency to a certain extent. However, the number of guides is not the more the better. Increasing the width of the exit within a certain range can increase the capacity of the exit and effectively reduce the evacuation time. When the speed of the guide is 75% of the pedestrian speed, the evacuation efficiency could be the highest. The model in this paper combines the advantages of small computation of cellular automata model and high clustering accuracy of Gaussian mixture model, which provides a new idea for pedestrian dynamics modeling and puts forward effective suggestions for pedestrian evacuation behaviors.

Key words： cellular automata model; Gaussian mixture model; fuzzy logic method; EM algorithm; guide

参考文献

[1] Guan J B, Wang K H, Chen F Y. A cellular automaton model for evacuation flow using game theory[J]. Physica A: Statistical Mechanics and its Applications, 2016, 461:655-661.
[2] Guo X W, Chen J Q, You S Z,et al. Modeling of pedestrian evacuation under fire emergency based on an extended heterogeneous lattice gas model[J]. Physica A: Statistical Mechanics and Its Applications, 2013, 392(9):1994-2006.
[3] Wang Q L, Dong H R, Ning B, et al. Two-time-scale hybrid traffic models for pedestrian crowds[J].IEEE Transactions on Intelligent Transportation Systems, 2018,19(11):3449 - 3460.
[4] 李世威,牛惠民.一种基于方向模糊可视域的行人流疏散仿真研究[J].交通运输系统工程与信息,2015,15(2):88-95.
Li Shiwei, Niu Huimin. Simulation of pedestrian flow evacuation based on direction fuzzy visual field[J].Journal of Transportation Systems Engineering and Information Technology, 2015, 15(2):88-95.
[5] Piccoli B, Tosin A. Time-evolving measures and macroscopic modeling of pedestrian flow[J]. Archive for Rational Mechanics and Analysis, 2011, 199:707-738.
[6] Yang X X, Yang X L, Wang Q L. Pedestrian evacuation under guides in a multiple-exit room via the fuzzy logic method[J]. Communications in Nonlinear Science and Numerical Simulation, 2020, 83:105138.
[7] Jiang Y Q, Zhang P.Modeling and simulation of pedestrian flow through hydrodynamics[J]. Procedia Engineering, 2012, 31:1039-1044.
[8] YangX X, DongH R, YaoX M, et al. Necessity of guides in pedestrian emergency evacuation[J]. Physica A: Statistical Mechanics and Its Applications, 2016, 442:397-408.
[9] Zhang P, Li X Y, Deng H Y, et al. Potential field cellular automata model for overcrowded pedestrian Flow[J].Transportmetrica A: Transport Science, 2020, 16(3):749-775.
[10] Yang X X, Xue S Q, Zhang J H, et al. The effect of waiting area design at the metro platform on passengers' alighting and boarding behaviors[J]. Applied Mathematics and Computation, 2019, 358:177-193.
[11] Shi X M, Ye Z R, Shiwakoti N, et al. A state-of-the-art review on empirical data collection for external governed pedestrians complex movement[J]. Journal of Advanced Transportation, 2018, 2018:1-42.
[12] Zhou D, Hu J, Gao X L, et al. Queuing behaviors at exit in cellular automaton model with S-queue[J]. International Journal of Modern Physics B, 2019, 33(8):1950064.
[13] Yang X X, Dong H R, Wang Q L, et al. Guided crowd dynamics via modified social force model[J]. Physica A: Statistical Mechanics and Its Applications, 2014, 411:63-73.
[14] Song X, Sun J H, Xie H N, et al. Characteristic time based social force model improvement and exit assignment strategy for pedestrian evacuation[J]. Physica A: Statistical Mechanics and Its Applications, 2018, 505:530-548.
[15] Yang X X, Yang X L, Wang Q L, et al. Guide optimization in pedestrian emergency evacuation[J]. Applied Mathematics and Computation, 2020, 365:124711.
[16] Kang Z X, Zhang L, Li K. An improved social force model for pedestrian dynamics in shipwrecks[J]. Applied Mathematics and Computation, 2019, 348:355-362.
[17] Wiler J L, Bolandifar E, Griffey R T, et al. An emergency department patient flow model based on queuing theory principles[J]. Academic Emergency Medicine, 2013, 20(9):939-946.
[18] Li Y L, Gao X, Xu Z W, et al. Network-based queuing model for simulating passenger throughput at an airport security checkpoint[J]. Journal of Air Transport Management, 2018, 66:13-24.
[19] Hu J, You L,Zhang H, et al. Study on queuing behavior in pedestrian evacuation by extended cellular automata model[J]. Physica A: Statistical Mechanics and Its Applications, 2017, 489:112-127.
[20] Li X L, Guo F, Kuang H, et al. An extended cost potential field cellular automaton model for pedestrian evacuation considering the restriction of visual field[J]. Physica A: Statistical Mechanics and Its Applications, 2019, 515:47-56.
[21] Ji J W, Lu L G, Jin Z H, et al. A cellular automata model for high-density crowd evacuation using triangle grids[J]. Physica A: Statistical Mechanics and its Applications, 2018, 509:1034-1045.
[22] Song W G, Yu Y F,Wang B H, et al. Evacuation behaviors at exit in CA model with force essentials: A comparison with social force model[J]. Physica A: Statistical Mechanics and its Applications, 2006, 371(2):658-666.
[23] Varas A,Cornejo M D, Mainemer D, et al. Cellular automaton model for evacuation process with obstacles[J]. Physica A: Statistical Mechanics and Its Applications, 2007, 382(2):631- 642.
[24] Nowak S, Schadschneider A. Quantitative analysis of pedestrian counterflow in a cellular automaton model[J]. Physical review E, 2012, 85(6):066128.
[25] Tang T Q, Shao Y X, Chen L. Modeling pedestrian movement at the hall of high-speed railway station during the check-in process[J]. Physica A: Statistical Mechanics and Its Applications, 2017, 467:157-166.
[26] Li Y X, Wu W J, Guo X, et al. Effect of group behavior on pedestrian choice for vertical walking facilities[J]. International Journal of Modern Physics B, 2020, 34(7):2050056.
[27] Hou L, Liu J G, Pan X, et al. A social force evacuation model with the leadership effect[J]. Physica A: Statistical Mechanics and its Applications, 2014, 400:93-99.
[28] 李军,吴茜蒙.紧急疏散下的人员行为及建模仿真[J].安全,2015,36(8):13-16.
Li Jun, Wu Qianmeng. Personnel behavior and modeling simulation under emergency evacuation[J].Safety and Security, 2015,36(8):13-16.
[29] Ma Y, Richard K K Y, Lee E W M. Effective leadership for crowd evacuation[J]. Physica A: Statistical Mechanics and Its Applications, 2016, 450:333-341.
[30] Zhang H, Liu H, Qin X, et al. Modified two-layer social force model for emergency earthquakeevacuation[J]. Physica A: Statistical Mechanics and its Applications, 2018, 492:1107-1119.
[31] 柴五一, 杨丰, 袁绍锋,等. 用于图像分割的多分类高斯混合模型和基于邻域信息的高斯混合模型[J]. 计算机科学, 2017, 45(11):272-278.
Chai Wuyi, Yang Feng, Yuan Shaofeng, et al. Multi-class Gaussian mixture model for image segmentation and Gaussian mixture model based on neighborhood information[J]. Computer Science, 2017,45(11):272-278.
[32] 欧丰林,林淑彬.融合高斯混合模型和深度学习的目标跟踪[J].吉林师范大学学报(自然科学版),2020,41(1):127-134.
Ou Fenglin, Lin Shubin. Fusion gaussian mixture model and deep learning target tracking[J]. Journal of Jilin Normal University (Natural Science Edition), 2020,41(1):127-134.
[33] 刘翔羽.基于混合高斯的行人检测模型[J].黑龙江工业学院学报(综合版),2019,19(12):55-61.
Liu Xiangyu. Pedestrian detection model based on mixed Gaussian[J]. Journal of Heilongjiang Institute of Technology (Comprehensive Edition), 2019,19(12):55-61.
[34] Fu L B, Song W G,Lo S M. A fuzzy-theory-based behavioral model for studying pedestrian evacuation from a single-exit room[J]. Physics Letter A, 2016, 380(34):2619-2627.
[35] Liu S B, Yang L Z,Fang T Y, et al. Evacuation from a classroom considering the occupant density around exits[J]. Physica A: Statistical Mechanics and Its Applications, 2009, 388(9):1921-1928.

Options

文章导航

模态框（Modal）标题

摘要

本文引用格式

Abstract

参考文献