A Data-Driven Car-Following Model Based on the Random Forest

The car-following models are the research basis of traffic flow theory and microscopic simulation. Among previous work, theory-driven dominant, while data-driven ones relatively rare. In recent years, related technologies Intelligent Transportation System (ITS) re- presented by Vehicles to Everything (V2X) technology have been developing rapidly. Utilizing ITS, large-scale vehicle trajectory data with high quality can be acquired, which provides foundation for modeling behavior based on methods. According this point, a model Random Forest (RF) method was constructed in Next Generation Simulation (NGSIM) dataset used calibrate train model. Artificial Neural Network (ANN) model, GM Full Velocity Difference (FVD) em- ployed comparatively verify proposed results suggest that work accurately describe car- following better performance under multiple indicators.