High-Order Computational Scheme for a Dynamic Continuum Model for Bi-Directional Pedestrian Flows

Abstract

In this article, we present a high-order weighted essentially non-oscillatory (WENO) scheme, coupled with a high-order fast sweeping method, for solving a dynamic continuum model for bi-directional pedestrian flows. We first review the dynamic continuum model for bi-directional pedestrian flows. This model is composed of a coupled system of a conservation law and an Eikonal equation. Then we present the first-order Lax-Friedrichs difference scheme with firstorder Euler forward time discretization, the third-order WENO scheme with third-order total variation diminishing (TVD) Runge-Kutta time discretization, and the fast sweeping method, and demonstrate how to apply them to the model under study. We present a comparison of the numerical results of the model from the first-order and high-order methods, and conclude that the high-order method is more efficient than the first-order one, and they both converge to the same solution of the physical model.

Reformulating the Hoogendoorn–Bovy predictive dynamic user-optimal model in continuum space with anisotropic condition

“…In contrast, the continuum modeling approach (Buckley, 1979;Wong, 1998;Wong et al, 1998;Xiong et al, 2011;Zhang et al, 2008) focuses on the overall behavior of travelers and is thus more suitable for macroscopic studies. It approximates a dense transportation system as a continuum in which users can move freely in a continuous two-dimensional space.…”

Integrating a statistical background-foreground extraction algorithm and SVM classifier for pedestrian detection and tracking

“…If Eqs (41)- (42) are introduced in the Langrangian function, a dual quadratic optimization problem is formed [28]:…”

“…It has been consolidating its influence in such fields as human-machine interaction, visual surveillance, traffic safety and national defense technologies, and most recently, behavior analysis of human group [42]. Due to its potential applicability and high complexity, pedestrian recognition has become a popular research topic in recent years.…”

Pedestrian detection in far infrared images

“…The applications are plenty and include freeway work zone analysis [1,27,36,37], automatic image search [34], human detection and modeling [12,69] and face recognition [7]. Object recognition in images has become a very important topic in the fields of traffic infrastructure and driving assistance system [30,54].…”

“…It is also useful in security, be it for automatic surveillance or people counting [63,69]. Those applications can be extended to low visibility conditions by using FIR images.…”