基於探偵數據之交通擁塞指標

Abstract

隨著市區道路使用者大幅增加，提供道路用者交通指標，可以幫助他們避開壅塞的地區節省旅行時間。由於具備GPS定位系統的車子逐漸普及，利用裝載GPS收發器的車子作為探針車取得即時交通資訊。TIC 可以整合這些GPS資訊以及歷史統計資料，進而估算出平均速度、旅行時間、交通壅塞指標。此篇論文提出D_queue和T_green這兩個壅塞指標來判斷當前的交通壅塞狀態，以及提出一個三步驟的演算法來估測各時段單一一個紅綠燈的的週期秒數(一個紅燈秒數+一個綠燈秒數)，結果顯示在於紅綠燈週期判斷的演算法其準確率可達80%以上，壅塞指標的演算法也可以作為一個判斷當前交通壅塞狀態的方法

Providing real-time Information, such as the travel time and traffic indicators, to road users and on-board navigation systems can save travel time and reduce fuel consumption. With the increasing popularity of vehicles equipped with GPS systems and wireless communication capability, these vehicles can be used as probe cars to collect the GPS information. A traffic information center (TIC) can integrate these GPS information and historical statistics into an average speed, travel time and traffic indicators. However, the effects of the traffic lights on these traffic information is significant in urban areas. We have to consider the effects of the traffic lights when we estimate the traffic condition. In this thesis, we use a stop-and-go model in analyzing the probes’ GPS trace to reduce the impact of traffic lights. We propose two traffic indicators, the queue length of a traffic light and the time when probes pass a traffic light, to determine the conditions of the traffic. In addition, we propose a three-steps algorithm to estimate the traffic light cycle time at an intersection. The experiment results indicate that the accuracy of the three-steps algorithm is higher than 80% and our traffic indicators algorithm is useful in determining whether the traffic is congested.