Connected and automated vehicles (CAVs) use vehicle-to-vehicle (V2V) wireless communication to drive in platoons, which can improve traffic efficiency and reduce energy consumption. However, some vehicles in a platoon may experience communication failures for various reasons, such as packet loss, signal blocking, or damage to vehicle communication module hardware. This will result in two phenomena: the platoon communication topology will be changed, and the behavior of vehicles without communication capabilities will degenerate to that of automated vehicles (AVs). The platoon will be unstable under the influence of these two phenomena. To mitigate the impact of these two phenomena on platoons, we propose a dynamic communication topology-based CAV-following model. This model takes into account changes in communication topology in a platoon due to communication failures or perceptual errors resulting from vehicle degradation. Two core features ensure that this model is able to perform platoon maintenance under communication failure. First, the CAVs in a platoon dynamically adjust the weight values of the acquired driving information based on the communication topology. Second, an adaptive Kalman filter (AKF) method is applied to reduce perceptual error. The effectiveness of the platoon maintenance scheme proposed in this study is verified by a series of simulation experiments. The results show that the proposed model can effectively maintain CAV platoons regardless of communication failures under steady-state or non-steady-state traffic flows. This finding indicates that CAV designers should design control schemes that take into account possible changes in the communication topology. Furthermore, our method has potential advantages for improving the platoon stability of CAVs driving on roads where communication is poor.

联网汽车和自动驾驶汽车 (CAV) 使用车对车 (V2V) 无线通信来进行队列行驶，这可以提高交通效率并降低能源消耗。但是，车队中的某些车辆可能会由于各种原因而出现通信故障，例如丢包、信号阻塞或车辆通信模块硬件损坏。这将导致两种现象：排通信拓扑将发生变化，并且没有通信能力的车辆的行为将退化为自动车辆（AV）的行为。在这两种现象的影响下，排会变得不稳定。为了减轻这两种现象对排的影响，我们提出了一种基于动态通信拓扑的 CAV 跟随模型。该模型考虑了由于通信故障或车辆退化导致的感知错误导致的排中通信拓扑的变化。两大核心特性保证了该模型能够在通信故障的情况下进行排维护。首先，排中的 CAV 根据通信拓扑动态调整获取的驾驶信息的权重值。其次，应用自适应卡尔曼滤波器（AKF）方法来减少感知误差。通过一系列仿真实验验证了本研究提出的排维修方案的有效性。结果表明，无论在稳态或非稳态交通流下的通信故障，所提出的模型都能有效地维持 CAV 排。这一发现表明，CAV 设计人员应设计控制方案，考虑到通信拓扑中可能发生的变化。此外，我们的方法对于提高在通信较差的道路上行驶的 CAV 的排稳定性具有潜在优势。