應用有限偵測器推估公路車流量之研究

Abstract

在先進用路人資訊系統(Advanced Traveler Information Systems, ATIS)中，為了要給予用路人準確的資訊，以作為路徑、運具選擇之依據，路徑旅行時間的推估是一項重要的課題。但是由於預算的限制，在大部分的都市交通路網中，並不是每一個路段皆有設置偵測器。 本研究以省道為研究對象，利用路口週遭路段流量間的關係，以及流量守恆來建立一個可以利用設置在省道上少數的偵測器所偵測到的流量、佔有率等資訊推算此路徑中其他沒有偵測器之路段流量的模式。本模式中的假設個路段的轉向比以及偵測器的佈設位置為已知。 本研究以新竹市路網為測試對象，同時假設不同的偵測器數量與安裝地點。測試的結果顯示出本研究所提出的方法具體可行，同時對所推估的結果具有一定的正確性。

Advanced Traveler Information Systems (ATIS) is one of the Intelligent Transportation System’s (ITS) subsystems. Travel time estimation is a very important and basic issue of ATIS. Travelers need estimated travel time to plan trips effectively. In order to estimate accurate travel speed, we need sufficient data sources as input. The most common data sources are from the vehicle detectors (VD). However, installing sufficient number of VDs costs tremendous amount of budget, and that is almost impossible in practice. Thus, the issue we concern most is how to estimate traffic flows of each road segment of the network from limited number of VDs. If we have traffic flows of all road segments, we can further estimate travel time. This core of this research is to propose a flow conservation based method to estimate traffic flows with limited VD sources. We assume the locations of installed VDs and the rates of turning right/left are know as a priori. We verify this method by numerical tests using the real network of Hsinchu city. The testing results show that the proposed method is technically sound.