Modelling of two-lane car-following behaviour considering driver's visual effect

In this paper, a novel two-lane car-following model considering driver's visual effect is proposed. Changes in driver behaviour are directly stimulated from visual information and an improved time-to-collision (TTC) calculation is presented. Using TTC, stimuli are formulated for describing the influence of leading vehicles. Besides, the model is established based on the optimal velocity model framework. The stability of the proposed model is discussed by theoretical analysis and numerical simulation. Moreover, the influences of parameters and the comparison between models are investigated by simulations, which show that the proposed model can effectively describe the influences. Finally, the model is calibrated and verified by NGSIM trajectory data, which shows that the proposed model fitting effect can be improved by 7.78%, 44.96%, and 32.07% respectively compared with other three models. This study may provide a basis for the design of control strategies for future intelligent connected vehicles in multi-lane environments.

Automated summary

In this paper, a novel two-lane car-following model is proposed, considering the driver's visual effect. The driver's behavior changes are directly simulated based on visual information, and an improved time-to-collision (TTC) calculation is introduced. The model is formulated using the optimal velocity model framework, and its stability is analyzed through theoretical analysis and numerical simulation. The proposed model is validated and calibrated using NGSIM trajectory data, showing better fitting effect compared to other models. The findings of this study can contribute to the design of control strategies for future intelligent connected vehicles in multi-lane environments.