Social force basedf model for traffic simulation

Abstract

To simulate a heterogeneous traffic flow is a promising branch of the research of traffic flow modeling. Historically, the demand for this kind of models comes from developing countries in which mixed traffic flows are common; however, recently the demand also comes from developed countries where the concept of “shared space” becomes acceptable and popular. In order to make a progress in this field, the approaches based on macroscopic model, mesoscopic model, and microscopic are considered by various researches. For macroscopic model, the continuum model based on the porous flow is proposed. This model is able to describe the phenomenon of heterogeneous queues for cars and scooters. For mesoscopic model, several extensions on Cellular Automation (CA) model are proposed, and also focus on modeling the traffic flows with a mixture of cars and motorcycles. The microscopic traffic flow model has several natural advantages in modeling heterogeneous flow. First, the microscopic models have already been applied to both vehicles and pedestrians, therefore a lot of successful experience can be inherited from those researches; second, aiming at describing the individual behavior of traffic participants, microscopic model is suitable for describing the difference between diverse traffic participants.

This paper is based on the preceding work in [1] [2]. An approach for modeling vehicle behaviors on the 2-dimensional space is proposed. A vehicle dynamic model is employed to simulate the attributes of vehicles, and extended social force model is adopted to describe the influence of environments on the simulated vehicle. In addition, several typical behaviors are tested and discussed in order to show the feasibility, prospects, and limitation of this approach.