Citation

Han Y, Liu H, Li L. Transportmetrica A: Transp. Sci. 2021; ePub(ePub): ePub.

Copyright

(Copyright © 2021, Informa - Taylor and Francis Group)

DOI

10.1080/23249935.2021.1977867

PMID

unavailable

Abstract

It is difficult for pedestrians with limited vision to obtain comprehensive environmental information during evacuations, so they cannot always make a proper evacuation plan to escape. For this purpose, we analyze the relationship between intermediate states and evacuation efficiency and propose an evacuation guidance model to guide pedestrians to form a series of expected states, which can reduce the evacuation time caused by congestion. The model includes three major factors that affect the evacuation time: the length of routes, the congestion of routes, and the density of exits. Furthermore, we propose a method based on reinforcement learning to optimize the guidance model. Finally, we utilize reciprocal velocity obstacle technology to build an evacuation model to verify the results of our study. The simulations indicate that the optimized guidance model can make full use of all exits during an evacuation and shorten the total evacuation time.

Language: en

Keywords

Crowd evacuation; evacuation guidance; limited vision; reinforcement learning