基于无人驾驶车数据的快速路排队长度实时检测方法

Real-time Queue Length Detection Method of Expressway Based on Autonomous Vehicles Data

交通拥堵是造成城市交通延误和运输资源浪费的重要原因。为了实时有效的对交通拥堵现象进行检测及预测,本文搭建了一套基于无人驾驶车数据的快速路排队长度实时检测系统。针对交通拥堵中的交通流现象和特点,本文基于无人驾驶车采集数据的特点,应用交通波理论,提出基于斜率法和LWR交通流模型法的两种方法对交通波波速进行估计,进而对排队长度的实时检测,最后使用NGSIM数据集对该系统进行验证。结果表明,基于LWR交通流模型法的方法二估计结果更接近于真实值,且无人驾驶车比例越高,估计结果越准确。实验证明该系统具有实时性、高效性和准确性的优点。

Traffic congestion is an important cause of urban traffic delays and waste of transportation resources.In order to detect and predict the queue length in real time,a queue profile estimation system on freeways is investigated based on data collected by autonomous vehicles.By applying the traffic wave theory,two methods are proposed to estimate the shockwave speed according to the characteristics of data collected by autonomous vehicles and traffic flow theory,which is then used for the measurement of queue profile.A set of NGSIM data are utilized to evaluate the performance of the proposed system.The results show that the second method has a better performance.And the higher penetration of drive-less vehicle used,the more accurate the estimation result.The results demonstrate that the proposed system is applicable for real-time queue profile estimation.