An Improved Heterogeneous Traffic Flow Model on Freeways Considering the Influence of Multiple Leading and Following Vehicles

Abstract

This study investigates the operational characteristics of heterogeneous traffic flow on freeways comprising human-driven vehicles (HVs) and connected autonomous vehicles (CAVs) by developing an improved heterogeneous traffic flow model that integrates both car-following and lane-changing behaviors of HVs and CAVs. The car-following and lane-changing behaviors of HVs are modeled using the intelligent driver model (IDM) and the minimizing overall braking induced by the lane-changing (MOBIL) model. For CAV car-following behavior, the IDM is enhanced to incorporate complete state information from multiple leading vehicles. Furthermore, the MOBIL model is refined to develop an autonomous lane-changing model for CAVs, accounting for the influence of multiple following vehicles. Game theory models the competitive and cooperative interactions among multiple CAVs, leading to a cooperative lane-changing model. Simulations are conducted using MATLAB. The results demonstrate that, compared with existing combinations of the IDM and MOBIL models, the proposed model yields substantial improvements in traffic flow stability and safety. An increased CAV penetration rate further improves traffic capacity, with particularly pronounced effects observed when the CAV penetration rate exceeds 0.4. These findings provide valuable theoretical insights into research on traffic flow modeling, capacity analysis, and traffic management strategies.