A cellular automaton crowd tracking system for modelling evacuation processes

Abstract

Crowd safety and comfort in highly congested places not only depend on the design and the function of the place, but also on the behaviour of each individual. In this paper, an integrated evacuation system is described. The proposed system comprises three stages. The main stage includes an efficient computational tool based on Cellular Automata (CA) capable of simulating main features of pedestrian dynamics during the evacuation of large areas, supported by a multi-parameterised graphical-user interface (GUI). Moreover, an image-processing tracking algorithm is used for the calibration of the system providing all the necessary information about the number of individuals and their distribution in the under test area. Finally, the VLSI implementation of the proposed model is straightforward due to the simplicity of the CA rule, thus leading to the design of a dedicated processor. © Springer-Verlag Berlin Heidelberg 2006.