Agent-based modeling: A case study in urban traffic simulation system

Modeling is one of the main methods researchers use to research our world. Different kinds of models can be settled to solve different problems, such as physical models, mathematical models, figural models etc. The main shortcoming of such traditional models is that they treat the members of the system as ones without aims, preferences and activities. Agent technology provides a new idea and methodology for modeling. Agent-based modeling pays more attention to the behavior and the interaction of the individuals in the systems, and the macroscopic characteristics of the system are the interactive result of all the individuals. In this paper, the agent-based modeling method is applied to the urban traffic simulation system. The process of the modeling is discussed, and a prototype of a simulation system is programmed. A reinforcement learning control method is tested on the prototype to verify the effectiveness of the modeling method proposed in this paper.