Car-following Model Based On Modified Optimal Velocity Function And Multiple Proceeding Vehicles

Car-following theory is an important part of the microcosmic traffic theory and the theoretical basis of the macroscopic traffic phenomena, which has great realistic significance in the traffic control and management. With the development of intelligent traffic system and vehicle equipment, drivers can obtain much more multiple and accurate information about road, traffic and environment, which may influence the way of the driving. Therefore, study of car-following model with the consideration of multiple preceding vehicles is of significant importance.Firstly, field data of car-following is gathered by vehicles equipped with high precision GPS. The car-following process is divided into three states based on the change of acceleration, they are start state, steady following state and slowed to a stop state, and change rules of velocity and acceleration of each state is analyzed. Numerical relationship between headway, velocity, acceleration and relative velocity is analyzed, the result shows that the exponential curve can fit the splashes of headway-velocity well and there is a strong linear relationship between the acceleration and related velocity.Then, the modified car-following model with the consideration of multiple preceding vehicles and driving behaviors is established. The optimal velocity function is modified based on the field data and the car-following model is modified with the consideration of multiple preceding vehicles’ driving behaviors. The stability solution of the model is gained by the mathematical and physical method. Parameters in the modified optimal velocity function and the modified car-following model is calibrated with the field data.Finally, the stability of the modified car-following model is analyzed by theoretical derivation and numerical simulation. Result of the theoretical derivation shows that the modified optimal velocity can fit the field data well. Result of the numerical simulation shows that the modified car-following model can describe the changing tend of the state and stop states, and shows higher stability under the disturbance caused by multiple driving behaviors.