Modeling and stability analysis of cyberattack effects on heterogeneous intelligent traffic flow

Abstract

The platooning of connected automated vehicles (CAVs) will have a dramatic impact on the future market. The open-access environment is available to share travel information among CAVs to format vehicular platooning, but makes cooperative vehicle platoons vulnerable to suffer cyberattacks. Cyberattacks are classified as bogus messages, replay/delay, and collusion attacks and affect the transmission of information on CAVs. Besides, heterogeneous traffic composed of different vehicles will be an extremely common form of traffic, including mixed operations of different models such as cars and trucks. Meanwhile, different car-following combinations influence the overall stability of heterogeneous traffic flow dynamics. This paper proposes a novel intelligent driving model considering cyberattacks and heterogeneous vehicles. In addition, this model integrates dynamic communication topology, which is a time-varying function determined by the communication range and the distance between vehicles. Through linear stability analysis, the stability conditions of the model are derived. On the other hand, numerical simulation was performed, and the results validated that under cyberattacks high proportion of cars and the information accepted from cooperative vehicles ahead can efficaciously enhance traffic stability and safety. Inversely, the proportion of car-following pairs and response/delay time will destabilize mixed traffic. From the perspective of platoon operation, the perniciousness of cyberattacks on CAV platoon is demonstrated, which lead to different levels of dangerous traffic behavior: unnecessary delay, congestions and potential collisions. The numerical results verified the feasibility and effectiveness of the model.