Modeling Cooperative Driving of Heterogeneous Vehicles in Tunnel Entrance Zones under Communication Quality Constraints

With the development of connected and automated vehicles (CAVs), the coexistence of CAVs and human-driven vehicles (HVs) has become commonplace. The unique environmental conditions at tunnel entrances, such as sudden changes in lighting, speed limits, road alignment contraction, and communication quality degradation—combined with the differences in perception and communication capabilities among heterogeneous vehicles, lead to complex evolutions in vehicle behavior and traffic conditions. Existing research mostly focuses on single disturbances or fixed communication topologies, making it difficult to systematically characterize vehicle cooperation mechanisms under multiple disturbances. Therefore, this paper proposes a modeling method that comprehensively considers the impact of environmental disturbances and communication quality. By analyzing the influence of speed limit and lighting changes on vehicle behavior, a “tunnel virtual force” model is proposed, thereby establishing an HV dynamics model that accounts for environmental disturbances. Simultaneously, for CAVs, a dynamic weight allocation mechanism based on communication quality indicators (signal-to-noise ratio (SNR) and packet loss rate (PLR))is designed. This enables real-time adjustment of the weights assigned to multiple preceding vehicles when communication quality fluctuates, thereby establishing a CAV dynamics model that more accurately characterizes the cooperative behavior of heterogeneous vehicles. By integrating heterogeneous information interaction topologies, this paper further constructs a unified hybrid fleet model to characterize the different information interaction and coordination mechanisms among HVs, CAVs, and autonomous vehicles (AVs). Here, AVs refer to CAVs that have lost their vehicle-to-everything (V2X) communication capabilities and have degraded to relying solely on onboard sensors; this degradation state consequently alters the information topology. The stability of the fleet model is analyzed using Lyapunov-Krasovskii theory, and simulation experiments are conducted to analyze the impact of communication quality variations on fleet stability and overall performance. Experimental results show that increasing CAV penetration and optimizing their spatial distribution can significantly improve fleet stability and traffic efficiency. Furthermore, the proposed dynamic weighting mechanism exhibits stronger adaptability and reliability than fixed weighting, significantly mitigating the coordination failures caused by communication quality degradation—including scenarios where CAVs degrade into AVs—and the resultant increase in energy consumption.