A Holonification Model Using Random Forest for Crowd Simulation

摘要

This paper proposes a new holonification method for data-driven crowd modelling using crowd data and machine learning methods. Using real-world data, we attempt to model the rules of joining each agent to a Holon and leaving it with random forests. Then we use this model in simulation. Also, because we use data from a specific environment, we test the model in another environment. The result shows that the rules derived from the first environment exist in the second one. It confirms the generality of the proposed method. Moreover, Because of using random forests, our proposed model is very fast and it can be used in a real-time simulation. Finally, this holonification model is applied to a crowd simulation. The evaluation result shows that the proposed method provides more realistic results.