Abstract Real-time queue length estimation and prediction provides useful information for proactively managing transportation networks. Queue spillback from off-ramps onto main lanes of freeways is one of the traffic issues caused by vehicular queues that can be efficiently managed using dynamic queue information. In this paper, a case-based reasoning algorithm combined with a Kalman filter is developed to provide real-time queue length estimations and predictions on freeway off-ramps. The estimations are based on occupancy readings from three loop detectors installed on a ramp. The proposed method is examined using a micro-simulation model of an off-ramp with a length of 650 meters and a traffic signal downstream of the ramp. The simulation results show an accuracy of ±3.15 vehicles in the queue in 60-second time intervals. In addition, a rigorous sensitivity analysis is conducted to examine the performance of the algorithm under various demand loading scenarios, time intervals, number of detectors used, and errors in prior estimations. The results show that the model performs well in terms of estimating and predicting the length of long queues on freeway off-ramps at various congestion levels. The outcomes of this study can be utilized to activate dynamic, responsive, and proactive queue management and traffic control measures.

中文翻译：

---

高速公路长匝道排队长度估计与预测模型

摘要 实时队列长度估计和预测为主动管理交通网络提供了有用的信息。从出口匝道到高速公路主车道的队列溢出是由车辆队列引起的交通问题之一，可以使用动态队列信息进行有效管理。在本文中，开发了一种结合卡尔曼滤波器的基于案例的推理算法，以提供高速公路出口匝道的实时队列长度估计和预测。估计值基于安装在斜坡上的三个环路检测器的占用读数。使用长度为 650 米的出口匝道和匝道下游的交通信号的微观仿真模型来检查所提出的方法。仿真结果显示，在 60 秒的时间间隔内，队列中的准确度为 ±3.15 辆车。此外，进行了严格的敏感性分析，以检查算法在各种需求加载场景、时间间隔、使用的检测器数量以及先前估计中的错误下的性能。结果表明，该模型在估计和预测不同拥堵级别的高速公路出口匝道上的长队长度方面表现良好。这项研究的结果可用于激活动态、响​​应式和主动的队列管理和交通控制措施。