Developing a car-following model with consideration of driver's behavior based on an Adaptive Neuro-Fuzzy Inference System

Based on a car following experiment with a synchronized fifth-wheel system, the Adaptive Neuro-Fuzzy Inference System (ANFIS) is utilized to develop a car following model. Wavelet analysis was used to denoise the high frequently collected fluctuating data. Drivers' reaction time was calculated using Cross Correlation analysis. In the car-following model developed, relative speed, distance headway between leading and following vehicle, and speed of following vehicle are the three input parameters; while acceleration of the following vehicle is the output. Comparing with the traditional General Motor (GM) car-following model, the results show that the ANFIS-based car-following model can better describe the freeway driving behavior.