Modeling bi-direction pedestrian flow by cellular automata and complex network theories

In this paper the experiment to study the features of pedestrians' walking preference is designed. Then the cellular automata model which considering pedestrians' walking preference features is built, in which the forward-parameter, right-parameter, surpass-parameter and the correction-parameters are included to mend the probability of the pedestrian getting to each neighboring cell. Based on this model and k-nearest-neighbor interaction pattern, the complex network of pedestrians is modeled. The simulation results obtained from the model well illustrate the density-speed curve and density-volume curve. Meanwhile the self-organization phenomena of the bi-direction pedestrian flow can be observed from the model simulation. In the further analysis of the pedestrian flow's basic parameter and the main feature parameters of pedestrians' complex network, it is found out that the average speed and the average path length are changed with the state of the flow. Finally it can be concluded that there is a linear negative correlation between these two parameters by fitting the data; in other words, pedestrian flow with shorter average-path length has a higher average speed.