Department of Civil and Environmental Engineering, Seoul National University, Seoul, Republic of Korea.
Department of Civil and Environmental Engineering and Institute of Construction and Environmental Engineering, Seoul National University, Seoul, Republic of Korea.
PLoS One. 2020 Dec 21;15(12):e0244329. doi: 10.1371/journal.pone.0244329. eCollection 2020.
Backward-moving kinematic waves (KWs) (e.g., stop-and-go traffic conditions and a shock wave) cause unsafe driving conditions, decreases in the capacities of freeways, and increased travel time. In this paper, a sequential filtering method is proposed to detect KWs using data collected in a connected environment, which can aid in developing a traffic control strategy for connected vehicles to stop or dampen the propagation of these KWs. The proposed method filters out random fluctuation in the data using ensemble empirical mode decomposition that considers the spectral features of KWs. Then, the spatial movements of KWs are considered using cross-correlation to identify potential candidate KWs. Asynchronous changes in the denoised flow and speed are used to evaluate candidate KWs using logistic regression to identify the KWs from localized reductions in speed that are not propagated upstream. The findings from an empirical evaluation of the proposed method showed strong promise for detecting KWs using data in a connected environment, even at 30% of the market penetration rates. This paper also addresses how data resolution of the connected environment affects the performance in detecting KWs.
后退运动运动波（KWs）（例如，停停走走的交通状况和冲击波）会导致驾驶不安全、高速公路通行能力下降和旅行时间增加。在本文中，提出了一种顺序滤波方法，用于使用在连接环境中收集的数据检测 KWs，这有助于为连接车辆开发交通控制策略，以阻止或减缓这些 KWs 的传播。该方法使用集合经验模态分解来过滤数据中的随机波动，该分解考虑了 KWs 的频谱特征。然后，使用互相关来考虑 KWs 的空间运动，以识别潜在的候选 KWs。使用逻辑回归来评估经去噪的流量和速度的异步变化，以从速度的局部降低中识别 KWs，这些降低不会向上游传播。对所提出方法的实证评估结果表明，即使在 30%的市场渗透率下，使用连接环境中的数据检测 KWs 具有很大的潜力。本文还讨论了连接环境的数据分辨率如何影响检测 KWs 的性能。
