Modeling and analysis of car-following models incorporating multiple lead vehicles and acceleration information in heterogeneous traffic flow

Abstract

In the foreseeable future, a complex heterogeneous traffic environment will emerge as Connected and Autonomous Vehicles (CAVs) coexist with Human-driven Vehicles (HDVs). Consequently, understanding the impact of CAVs on car-following behavior and the operational characteristics of heterogeneous traffic flow becomes crucial before their widespread deployment. To tackle this challenge, this research proposes an improved car-following model based on the Intelligent Driver Model (IDM). The model incorporates the position, velocity, and acceleration information of both front and sub-front vehicles in the heterogeneous traffic flow. The impact of different types of information on the model's stability is verified through linear stability analysis while investigating the operational characteristics of traffic flow during vehicle start-up. Additionally, the car-following modes are classified based on the type of leading vehicle, and the corresponding following model is formulated. The results indicate that the improved model significantly improves traffic flow stability, particularly when considering acceleration information. Compared to the front vehicle, the influence of the sub-front vehicle on traffic flow stability is less significant, but their combined impact yields positive effects. Furthermore, the improved model reduces the start-up time of vehicles at signalized intersections by 7.9% and enables a smoother start-up process for vehicles. Moreover, CAVs can mitigate the impact of HDV's disturbances on the overall heterogeneous fleet operation by adjusting their spacing relative to the front vehicle. With an increasing penetration rate, the velocity fluctuation of the entire fleet decreases.