The vehicle lane changing behavior is difficult to identify and predict due to the influence of traffic environment then, a learning model based on support vector machine was proposed to simulate the behavior decision of the driver on a freeway. By analyzing the characteristics and laws of the vehicle in the phase of lane changing, the appropriate physical quantity was chosen as input parameter of model. Based on the NGSIM database, a proper method was used to extract the samples, and the sample data was processed by differential filtering, Kal...
