Experimental Study and Modelling of Pedestrian Space Occupation and Motion Pattern in a Real World Environment

Abstract

In this work we tackle the statistical study and modelling of the usage of space by pedestrians in a real world environment. A large amount of pedestrian trajectories is collected in a corridor used just as a transition place, and the density and velocity distributions are analysed as functions of the distance from the walls. The empirical data are fitted to a model assuming the density and velocity to be determined through a Boltzmann factor by a comfort function depending on the distance from the walls and assuming a maximum on the left side of the corridor (Japanese traffic convention). The empirical data are then compared to numerical simulations using pure collision avoidance models, to better analyse the influence of the environment on the pedestrian distribution and to investigate how to introduce in collision avoiding the bias that makes people walk preferentially on a given side of a corridor.