影响因子：7.6
DOI码：10.1016/j.trc.2025.105116
所属单位：Hefei University of Technology
教研室：School of Automotive and Transportation Engineerin
发表刊物：Transportation Research Part C
刊物所在地：UK
项目来源：National Natural Science Foundation of China
摘要：In the connected vehicle environment, cyberattacks on the communication of connected vehicles can change traffic flow dynamics, which can severely degrade urban traffic system performance. Most studies have been carried out to develop a series of measures (e.g., attack detection, message encryption) to respond to cyberattacks. Notably, developing a defensive control strategy is also necessary against cyberattacks and can enhance the robustness of a traffic system. Unlike existing microlevel studies (single-vehicle or vehicular platoons), this paper focuses on the risk of regional cyberattacks on future traffic systems resulting in large-scale congestion due to unbalanced vehicle distributions. First, macroscopic fundamental diagram (MFD) theory is introduced to analyse the evolutionary mechanism of the impact of cyberattacks on road network performance. Second, on the basis of the analysis results, a two-layer control framework is developed to respond to cyberattacks on large-scale road networks. The upper layer is designed as a perimeter control (PC) to regulate the traffic inputs to the network affected by the attack (protected network); the lower layer is the distributed max pressure (MP) control, which incorporates a historical experienced data-based MP node selection algorithm under cyberattacks to improve the robustness of network performance. Finally, the proposed control framework is tested in a predefined 12×6 grid network. The results show that our proposed two-layer control framework improves the robustness of the network system compared with two-layer control, which combines other classic control methods (Bang-Bang, PI) and significantly reduces the total vehicle travel time and optimizes the traffic density distribution on the road network.
论文类型：期刊论文
学科门类：工学
文献类型：J
卷号：174
页面范围：105116
字数：14631
是否译文：否
发表时间：2025-04-03
收录刊物：SCI、EI
发布期刊链接：https://www.sciencedirect.com/science/article/pii/S0968090X25001202