A Cellular Automaton Model for Crowd Evacuation and Its Auto-Defined Obstacle Avoidance Attribute

In this paper, a crowd evacuation model based on Cellular Automata (CA) is described. The model takes advantage of the inherent ability of CA to represent sufficiently phenomena of arbitrary complexity and to be simulated precisely by digital computers as well. Pedestrian movement depends on their distance from the closest exit, which is defined dynamically. The adoption of Manhattan distance as the reference metric provides calculation simplicity, computational speed and improves significantly computational performance. Moreover, the model applies an efficient method to overcome obstacles. The latter is based on the generation of a virtual field along obstacles. A pedestrian moves along the axis of the obstacle towards the direction that the field increases its values, leading her/him to avoid the obstacle effectively. Distinct features of crowd dynamics and measurements on different distributions of pedestrians have been used to evaluate the response of the model.