利用GPS轨迹的转向级交通拥堵精细分析

目前,基于GPS轨迹探测城市交通状态的研究缺乏对不同行驶方向的交通拥堵精细感知。针对此问题,本文提出一种基于出租车GPS轨迹的转向级交通拥堵事件探测方法。该方法首先在分析出租车运营行为特征的基础上,采用特征聚类方法滤选出能够反映真实交通状态的有效轨迹段;然后基于滤选后轨迹分析当前道路交通运行状态,探测城市路网中轻度、中度、重度3种不同强度的交通拥堵事件;最后基于拥堵事件轨迹分析交叉口不同转向的拥堵时间、拥堵强度和拥堵距离等转向级精细交通拥堵状态。试验结果表明,本文方法不仅能有效探测路网中不同强度的拥堵事件,而且能实现交叉口转向级拥堵事件的精细分析。

Fine-grained analysis of traffic congestions at the turning level using GPS traces

For the issue that existing approaches on studying traffic conditions using GPS traces lack of detailed analysis of traffic congestion, this paper puts forward an approach for detecting traffic congestion events based on taxis' GPS traces at turning level. Firstly, this approach analyzed taxis' operating patterns and filtered valid traces. Then this approach detected traffic congestion traces of three different intensities:mild congestion, moderate congestion and serious congestion, based on analyzing traffic conditions from the filtered valid trace segments. Finally, traffic flow speed, congestion time and congestion distance of each turning direction at an intersection were explored at a fine-grained level. The experimental results show that the proposed approach is able to detect congestions of different intensities and analyze congestion events at turning level.