Velocity-repulsion model for pedestrian dynamics based on a multi-state modelling framework. (1st January 2023)

• Record Type: Journal Article
• Authors: Ding, Heng; Chen, Juan; Ma, Jian; Wang, Qiao; Shi, Congling
• Is Part Of: Transportmetrica. Volume 19:Number 1(2023)
• Publication Date: 2023-01-01
• Subjects: Pedestrian dynamics -- multi-state modelling framework -- velocity-repulsion model -- velocity-density relation -- velocity-headway relation -- flow-width relations -- self-organisation phenomena; Transportation -- Periodicals; Transportation -- Research -- Periodicals; 388.072
• DOI: 10.1080/23249935.2021.2017065
• Languages: English
• ISSNs: 2324-9935

Abstract

Abstract : In this work, we proposed a multi-state modelling framework that transforms the pedestrian simulation problem into the design of five modules, namely pedestrian state, environmental information, movement action, action function, and state transition rules. Based on the framework, a velocity-repulsion model is constructed with the feature that the faster a pedestrian walks, the greater distance he will keep from others. The model sets several states with different velocity and repulsion parameter couples for pedestrians who take action based on the current state and adjust the velocity and repulsion distance by updating the state. Simulation works are performed for validation. The simulated velocity-density and flow-width relations fit well with experimental data. Two characteristic forms of velocity-headway relations for acceleration and deceleration are revealed in the simulation and confirmed in the experimental data. Self-organisation phenomena of the bottleneck scenarios are reproduced.