標題: 使用高斯狀態空間模型與旅行時間估計旅次起迄
Estimation of Dynamic Origin-Destination by Gaussian State-Space model with travel time
作者: 張志浩
Chih-Hao Chang
周幼珍
Yow-Jen Jou
統計學研究所
關鍵字: 卡門濾波模型;狀態空間模型;Kalman Filter;State-Space model
公開日期: 2003
摘要: 中文摘要 動態旅次起迄推估長久以來為運輸管理之核心,經由路網中偵測器所收集的資料,可以進行路網相關交通狀態的估計或預測,並根據預測結果進而模擬短時間內的交通狀況擬定適當的交通控制與管理方式,維持交通順暢。本研究為對高速公路旅次起迄流量之估計採用狀態空間模型並考慮兩地之間的旅行時間,配合上統計理論上的卡門濾波模式(Kalman filter)與吉柏司樣本法(Gibbs sampler)去構築本研究之模型,並比較傳統沒有考量旅行時間的模型之差異性。
ABSTRACT Estimation of dynamic estimation of the O-D flow is the kernel of the traffic management for a long time. Through the data collected by the detectors in the network, we can estimate and predict the traffic condition about the network. According to the prediction results we can simulate the traffic condition and draft the associated appropriate traffic control and management to keep the free traffic. In this thesis, using the state-space model with travel time, estimation of the O-D flow of the freeway is considered. Using the Kalman filter and Gibbs sampler to complete the revised model, we compare the results with these from the traditional model without considering the travel time.
URI: http://140.113.39.130/cdrfb3/record/nctu/#GT009126519
http://hdl.handle.net/11536/55557
Appears in Collections:Thesis


Files in This Item:

  1. 651901.pdf

If it is a zip file, please download the file and unzip it, then open index.html in a browser to view the full text content.