Citation

Feliciani C, Nishinari K. Transportmetrica B: Transp. Dyn. 2019; 7(1): 865-896.

Copyright

(Copyright © 2019, Hong Kong Society for Transportation Studies, Publisher Informa - Taylor and Francis Group)

DOI

10.1080/21680566.2018.1517061

PMID

unavailable

Abstract

In this study, an alternative method to estimate velocity and density in pedestrian crowds is presented. The proposed approach uses angular velocity and acceleration measurements obtained from inertial sensors to judge amount of body motion and therefore estimate walking velocity. Taking into consideration practical aspects, commercial devices (generic tablets and smartphones) are used to measure inertial quantities showing application potentials but also limitations in regard to accuracy. Several experiments have been performed in increasingly realistic scenarios to assess the suitability of the method for public pedestrian spaces.

RESULTS show that estimation for speed and density is very accurate when position and orientation of the devices are known. Estimations are still reliable in more heterogeneous situations employing different devices, with accuracy increasing with the number of pedestrians being surveyed. The proposed approach may help getting a general picture of pedestrian crowds' condition over large areas with greater employment flexibility.

Language: en

Keywords

crowd control; crowd properties; electronic devices; inertial sensor; Pedestrian traffic