以加入準確度之團隊共識法進行交通資料融合

Abstract

準確的即時交通資訊是提供交通管理者與用路人做決策重要的工具之一，即時路況來源有偵測器、探針車等，這些資料各有其使用範圍及限制，為了要得到更佳的即時路況資料，交通資訊管理者開始採用資料融合(Data Fusion)的方法，以得到較為可靠的交通資料。 團隊共識法由Chung, Shent跟Basir在1996年提出，2004年由吳欣潔修改後成為較適合交通資料融合之模式，稱為最佳權重法。本研究採用團隊共識法為基礎進行資料融合，加入吳欣潔及吳瑞豐改進後的模式，並在模式中加入準確度之量測，欲改善融合之結果。本研究以電腦模擬之資料進行測試，測試後發現加入準確度確實可改善融合結果，但使用上及實際上限制較多，尚有可改進之處。

Accurate real-time transportation information is one of the most important tool for traffic administrators and travelers. We can obtain real-time transportation information from VDs and probe vehicles. But these source has its own constrains. For a better and more reliable data, the transportation administrators started to use a data fusion method to obtain a more accurate data. Team consensus was proposed by Chung, Shent and Basir in 1996 and modified by Hsin-Chieh Wu in 2004. This research is based on team consensus and Hsin-Chieh Wu’ model. We add accuracy in the model to obtain a more reliable result. We use simulation data to evaluate the performance of our model. The result shows that adding accuracy could obtain a better performance, but there are some constrains when usin