Traffic Flow Phase Transition at Bottleneck Sections Based on Driving Feedback Delay and Connected Density Difference Feedback Under Cyber-Attack Conditions

In real driving environments, Connected Vehicles (CV) and Connected and Autonomous Vehicles (CAV) are susceptible to cyber-attacks, while Human Driven Vehicles (HDV) exhibit driving feedback delay. When driving feedback delay and cyber-attacks overlap, especially at traditional bottleneck sections, traffic congestion and disorder are further exacerbated. To address these issues, this paper proposes a lattice model for traffic flow at bottleneck sections based on driving feedback delay and connected density difference feedback. Firstly, the impacts of driving feedback delay and cyber-attacks on traffic flow are incorporated into the model to make it more realistic. Secondly, traffic flow models are established for two typical scenarios: two-lane traffic interruption bottleneck sections and signal-free roundabout bottleneck sections. Finally, CAV deploy connected density-difference feedback, boosting flow stability and jam resistance. The effectiveness of the model in optimizing traffic flow is verified through stability analysis and numerical simulations. Furthermore, the simulations of bottleneck scenarios indicate that the proposed model can effectively alleviate congestion and enhance traffic flow stability. This study provides a concise theoretical framework and actionable guidance for understanding traffic evolution and enhancing its resilience and reliability in connected environments.