一個以行動網路位置更新為基礎的車輛速度估計機制

Abstract

建置即時交通資訊服務系統為智慧型運輸系統(Intelligent Transportation System, ITS)中重要的一部分，藉由提供即時且完整的交通資訊，不論是行前或行進中的路況資訊，用路人可以選擇替代道路或不同的運具，有效地降低用路人的旅行成本，對改善交通擁擠甚有助益。相較於透過傳統的固定式車輛偵測裝置(Stationary Vehicle Detector, SVD)來取得即時交通資訊，需要花費龐大的架設及維護費用且具有高損壞率；透過無線通訊網路追蹤分析使用者的移動行為，這種機制不需要額外設備的建設，而且可以提供更充足、即時的交通資訊。因此，本研究提出利用手機的一般位置更新(Normal Location Updates, NLU)及定期位置更新(Periodic Location Update, PLU)兩種訊號為作為定位的基礎，以計算出目標路段即時車速，我們利用NLU來估計目標路段的車流量，並以PLU來估計車密度，透過車流量除以車密度求得車速。實驗結果顯示，與實際車速值比較，於一般情況、離峰及尖峰時刻時其車速預測準確度分別為87.13%、82.29%及86.34%，因此可有效估計目前的行車速度和提供即時交通流量，予以用路人參考。

In recent years, with the advance of technology and economic growth, Intelligent Transportation System (ITS) has become more and more important. The construction of real-time traffic information service is an important part of ITS. For a road user, knowing real-time traffic information would help him in choosing better roads avoid congestion areas. Traffic monitoring based on cellular network data, also referred to as Cellular Floating Vehicle Data (CFVD), can be more cost-effective than traditional approaches, such as vehicle detector, because no field installation or maintenance. Therefore, an estimation method for average traffic speed using Normal Location Update (NLU) and Periodic Location Update (PLU) is proposed in this study. In experiments, we compare the estimated traffic information from VD. The results show that the accuracy of traffic speed estimation is 87.13% in normal case, 82.29% and 86.34% in off- peak case as well as peak hour respectively, so it is feasible to estimate the traffic speed with higher efficiency and lower cost solution for real-time traffic information service.