基於WAVE/DSRC技術之高速公路電子收費系統

Abstract

隨著無線通訊網路技術的進步，電子元件價格漸漸下降，各國政府紛紛投入智慧型運輸系統 (Intelligent Transport Systems, ITS)的研究。高速公路電子收費系統(Electronic Toll Collection, ETC)為 ITS 重要的應用，其利用電子收費方式，將原本在收費站採取人工收費的過程自動化，具有節能減碳、 降低車輛零件損耗、改善交通狀況等優點。其中多車道自由流動(Multi-lane free-flow, MLFF) ETC 架構 不受閘道與車道的限制，允許車輛以較高的速度通過收費站或收費區，為目前各國的趨勢。 然而 MLFF ETC 系統需仰賴穩定通訊品質才能維持高流量的扣款交易，現有的專用短距通訊 (Dedicated Short Range Communication, DSRC)技術中並不都可滿足MLFF ETC 這樣的需求。而車用環 境無線存取(Wireless Access in Vehicular Environments, WAVE)技術為因應車用網路環境對於高速、低延 遲、且穩定的傳輸需求所制定的一套短距通訊標準，採用專屬於車用系統的網路頻帶，能夠做到在短 距離之內可靠更迅速的資料傳輸。有鑑於此，本計畫第一年將研發以WAVE/DSRC 為基礎MLFF ETC 系統，設法利用其多重通道的特性，提供高車流環境中穩定且快速的交易需求。 此外，MLFF ETC 系統允許車輛可任意切換車道，執法系統較難達成準確的扣款資料與車牌影像 匹配。因此，本計畫第二年將研發一套可靠的執法影像匹配系統，利用雙分圖比配對(Bipartite Graph Matching)方法，辨別未付費的車輛，並結合車用定位系統(Vehicle Positioning System, VPS)以及行車視 訊紀錄器(Video Driving Logger)，辦識車輛所在車道，只需對車道所在的影像進行匹配即可，可進一 步提升執法系統的準確性。

Owing to the dramatic cost down on electronic components and the advances on wireless technologies, the development of Intelligent Transportation System (ITS) has drawn intensive attention in recent years from many countries. Highway Electronic Toll Collection (ETC) is one of the most important applications in ITS, which replaces labor power by electronic components, providing many advantages. In particular, the Multi-lane Free-Flow (MLFF) architecture allows vehicles passing through tolling station in higher speed. However, MLFF ETC requires more reliable and scalable communication capability to handle intensive tolling transactions. The Wireless Access in Vehicular Environment (WAVE) is an emerging Dedicated Short Range Communication (DSRC) technique, which provides high speed, low latency, and reliable communication for vehicular environments. Besides, matching tolled records with photo data of license plates is not a trivial task in MLFF ETC, because vehicles may change their lane from time to time. In the first year of this project, we will investigate a WAVE/DSRC-based MLFF ETC system. The system will exploit the multi-channel feature in WAVE/DSRC to achieve fast and stable toll collections. In the second year, we will design a reliable toll-record-and-license-plate matching mechanism using the bipartite graph matching approach. Moreover, we will integrate the Vehicle Positioning System (VPS) with the Video Driving Logger to identify the lane on which each vehicle is located, which can greatly improve the precision of the matching system.