Human Behavior Modeling for Evacuation From Classroom Using Cellular Automata

Understanding and simulating human behavior during the evacuation is essential in cognitive computing, which can be used to guide pedestrians to evacuate quickly and to avoid a potential hazard. In this paper, we propose a model to simulate the human behavior for evacuation from a classroom. Specifically, first, we give an improved evacuation model based on cellular automata, which consists of the static floor field and the dynamic floor field. Then, we present the detailed algorithms for calculating both the floor fields. The static floor field is calculated using an A-star algorithm, which can be used to solve the evacuation environments with and without obstacles. The dynamic floor field is calculated based on the impact of the number of pedestrians in each exit area on evacuation. Finally, the calculation method of pedestrian movement probability and the model evolution rules are given. The simulation experiments using our model have been conducted, and comparisons between our model and some state-of-the-art evacuation models are also conducted. The experimental results show that the model is effective and can reproduce the evacuation experiments conducted by students well. The model is expected to yield the optimal evacuation plan from the indoor environment.