Macroscopic characteristics of road network traffic flow under delay cyberattacks in a connected vehicle environment

Connected Autonomous Vehicle (CAV) technology enables real-time path information provision and smaller headway distances, and is adopted to improve the efficiency and safety of road network traffic flow. However, dynamic information interaction between CAV and the platform is mainly transferred through communication networks. The open communication environment is vulnerable to cyberattacks, leading to security threats such as delays and interruptions in traffic networks. Existing studies on the impact of cyberattacks on traffic mainly focus on individual vehicles or queues, which cannot be used to analyse the effects of the cyberattacks on the macroscopic traffic network. To acquire the impact of information delay caused by cyberattacks on traffic networks, based on the macroscopic fundamental diagram (MFD) theory, this paper carries out two works. Firstly, the effect of cyberattacks with different durations on the MFD of traffic networks in a grid-connected environment is analysed using a classical grid road network and a real road network as the analysis objects. Secondly, the variability of the impact of cyberattacks on the MFD of road networks is analysed under different CAV penetration rates as well as under different cyberattack durations. The experimental results demonstrate three findings: (i) when the road network is in a congested flow state, the MFD curve of the traffic network is more affected by cyberattacks; (ii) different cyberattack delay times will vary depending on the size and complexity of the road network; and (iii) the impact degree of cyberattacks decreases the traveling completion flow of the traffic network, and which increases with the rise of CAV penetration.