Full metadata record
DC Field | Value | Language |
---|---|---|
dc.contributor.author | 何佳儒 | en_US |
dc.contributor.author | 王晉元 | en_US |
dc.contributor.author | Wang, Jin-Yuan | en_US |
dc.date.accessioned | 2015-11-26T01:07:51Z | - |
dc.date.available | 2015-11-26T01:07:51Z | - |
dc.date.issued | 2010 | en_US |
dc.identifier.uri | http://140.113.39.130/cdrfb3/record/nctu/#GT079732506 | en_US |
dc.identifier.uri | http://hdl.handle.net/11536/45374 | - |
dc.description.abstract | 先進大眾運輸系統(Advanced Public Transportation System, APTS)的公車到站時間預測能提供大眾運輸系統使用者有用的乘車資訊,降低民眾搭公車的等待時間,藉其提昇民眾搭乘公車意願,減少使用私人運具。在車隊管理方面,亦可提供業者公車到站時間,協助業者有效管理車隊。 市區公車的到站時間預測,需考量路口號誌及站牌上下客的停等時間,因此本研究以公車裝配之全球定位系統(Global Positioning System, GPS)回傳的即時資料為主要資料來源。 本研究先利用GPS資料建立歷史資料庫,在資料比對上採用k-NN法預測公車到站時間。其中,因為實際的路段資料可能會有遺漏的情形,將歐基里德距離除上可取得的路段資訊數量作調整,以避免因較多的路段資訊造成較大的誤差。在計算路段平均速度時,採用延遲加總的方法,以反映車輛在行進中所遭遇的交通狀況。最後以台北市區公車為實測對象,以隨車調查的實際到站時間作驗證,探討在不同情況本研究方法的表現績效。 就整體結果而言,本研究與顯著優於現有系統之預測表現,表示本研究之預測模式可用,且在少站位數及下午尖峰情境下,本演算法有比較好的表現。 | zh_TW |
dc.description.abstract | The estimation of urban bus arrival time using the GPS data is a fundamental function of Advanced Public Transit Systems (APTS). This is a difficult task due to the complications of traffic, signals, and bus stops. We first propose a database schema to store the collected historical data. Then, a k-NN based method is developed to derive the real time estimation of bus arrival time. The real world data collected from various routes in Taipei City is used for testing purpose. The testing results show that the proposed method performs better than the existing methods. The average error is within 1.5 minutes. | en_US |
dc.language.iso | zh_TW | en_US |
dc.subject | k-NN法 | zh_TW |
dc.subject | 市區公車 | zh_TW |
dc.subject | 到站時間預測 | zh_TW |
dc.subject | k-NN method | en_US |
dc.subject | urban bus | en_US |
dc.subject | arrival time prediction | en_US |
dc.title | 應用k-NN模式於市區公車到站時間預測之研究 | zh_TW |
dc.title | The Study of Urban Bus Arrival Time Prediction Using the k-NN Model | en_US |
dc.type | Thesis | en_US |
dc.contributor.department | 運輸與物流管理學系 | zh_TW |
Appears in Collections: | Thesis |
Files in This Item:
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.