Modeling and application analysis of car-following model with predictive headway variation

This work proposes an improved microscopic model by taking the predictive driving behavior of the preceding car into account. Firstly, we use the linear method to investigate performance of the proposed model, deduce the critical linear stability conditions and obtain a neutral curve accounting for stability of the new model. Then, using the nonlinear analysis technique, we derive the equations of Burgers, the Korteweg-de Vries (KdV) and the modified KdV (mKdV), respectively, to describe the triangular shock wave, soliton density wave and the kink-antikink density wave. Finally, numerical experiments unfold from two aspects: one for verifying the theoretical analysis results and another for testing the impact of predictive driving behavior of the preceding car on traffic stability. Simulation results show that the new model can stabilize traffic flow compared with the FVD model and confirm that the existence of three kinds of density waves in different phase diagram regions. (C) 2019 Elsevier B.V. All rights reserved.