本研究以模擬分析為基礎，構建固定式偵測器與移動式時空資料之動態轉換函數，並將兩種不同資料來源之時空一致化，進行旅行時間推估之資料融合運算。
本研究以一實際都市幹道為範圍，經模擬軟體(PARAMICS)模擬調查環境中固定式偵測器與移動式偵測器(即探針車)之行進軌跡並輸出模式運算所需交通資料與旅行時間真值。其中，固定式偵測器之資料運算，係以每時區(5分鐘)之佔有率所推算之路段密度為基礎經轉換函數運算得空間平均速度後以計算推估旅行時間；而以探針車資料為主之運算則以購建多輛個別之軌跡速率並加權之平均速率與真實平均旅時間之動態轉換。之後的融合運算則是運用經推導正規化之迴歸式後校估其對應之係數作為融合所需之加權係數。
本研究並以另加入不經過轉換函數處理之推估旅行時間基本模式與運用轉換函數推估旅行時間之模式進行比較。經以平均絕對誤差百分比(Mean Absolute Percent Error, MAPE)為指標之評量結果顯示，在移動式轉換函數模式部份，以探針車之軌跡點速度作為轉換函數之模式輸入項之旅行時間推估模式有良好的績效表現，其推估旅行時間校估之MAPE在1.96%到5.62%之間，模式驗證部分在2.77%到4.34%之間，整體平均為3.47%；而以固定式偵測器資料為基礎之轉換函數模式其旅行時間推估部分之MAPE在8.28%到26.98%之間; 模式驗證部分則在1.99%到26.94%之間，整體平均為16.45%；融合模式推估旅行時間校估部分在0.4%到10.43%之間，驗證部分在0.15%到9.27%之間，整體平均為4.52%。因此就本研究而言，以探針車資料為基礎之轉換函數模式具有相當優良之績效表現，而融合模式可有效平衡兩不同資料來源之旅行時間推估差異；此外相較於基本模式之MAPE值在6.77%到16.16%之間，運用轉換函數模式之旅行時間推估績效明顯較佳。

Dual-loop vehicle detectors and/or probe vehicle technique are two most commonly used devices to perform instantaneous (real time) traffic data collection for either traffic management or travel information.  In particular, travel time information is of the major importance which can not be obtained directly but derived from various models/algorithms using detected traffic flow related data as input.  

Transfer function methods with the capability to calibrate bivariate relations using time series data was recognized to capture the effect of dynamic nature real time data instead of the commonly used static or steady-state relation.  It is therefore the focus of this thesis to develop an integrated procedure of the above-mentioned two difference data sources in the application of estimation of travel time along a specific road segment.  Transfer functions were calibrated for traffic data from loop detector and probe vehicle data streams respectively for each unified space-time interval to recognize the inherent differences between them.  Finally, data fusion technique was applied to obtain final estimate of travel time by integrating both estimates from fixed loop detector and probe vehicle.  A basic model without applying transfer function technique was also constructed to serve as a benchmark, from which the credibility of transfer function was presented.

This study employed Paramics, a microscopic traffic simulation program, as primary research tool of data sources to evaluate the proposed issues.  Care has been taken to calibrate simulation parameters using true field survey data to ensue the consistency. 

Travel time estimations were performed and evaluated for four models: exclusive loop-detector data with transfer function, exclusive probe vehicle data with transfer function, fusing both data with transfer function, and fusing data without transfer function.  Mean Absolute Percent Error (MAPE) values were calculated as evaluation criteria compared with simulated average over all true travel times traced by all vehicle trajectories within the studied space-time.
The results showed that travel estimation by probe vehicle data with transfer function performs better than that estimated by vehicle detector data with transfer function. And the fusion model acted as balance role to bridge the deviation of travel times from both detector and probe vehicle models. Finally, comparing with base case, where no transfer functions applied, we demonstrated the quality of transfer function by significant improvement of MAPE values over those of the base model.

目錄
目錄	I
圖目錄	III
表目錄	V

第一章 緒論	1
1.1 研究動機與背景	1
1.1.1 研究背景	1
1.1.2 研究動機	1
1.2 研究目的與範圍	2
1.2.1 研究目的	2
1.2.2 研究範圍	2
1.3 問題界定	4
1.3.1 固定式與移動式偵測器資料形式與時空特性	4
1.3.2 旅行時間	5
1.4 研究方法	6
1.4.1 線性轉換函數模式	7
1.4.2 融合模式	7
1.4.3 Oh模式	7
1.5 研究流程	7
第二章 文獻回顧	9
2.1 旅行時間相關研究	9
2.1.1 固定式偵測器資料	9
2.1.2 移動式	15
2.2 資料融合	17
2.3 小結	22
第三章 模式理論	24
3.1 固定式與移動式旅行時間推估模式	24
3.1.1 線性轉換函式模式	24
3.1.2 模式輸入項取得方法	29
3.2 融合模式(加權平均法)	35
3.3 資料流向	40
3.3.1 固定式旅行時間推估模式	40
3.3.2移動式旅行時間推估模式	45
3.3.3 融合模式	47
第四章 模擬實驗路網之構建及資料蒐集	50
4.1 PARAMICS 特性與簡介	50
4.2路網構建與參數校估	55
4.2.1路網構建	55
4.2.2參數校估	56
4.3 資料蒐集	58
4.3.1 固定式資料取得	59
4.3.2 移動式資料取得	60
4.3.3 模擬真值取得	61
第五章 模式校估與驗證	63
5.1 旅行時間模式校估	63
5.1.1 固定式	63
5.1.2 移動式	68
5.1.3 融合模式	73
5.2 模式驗證	77
5.2.1 驗證方法與說明	77
5.2.2 路段旅行時間驗證結果	78
5.3 轉換函數模式與基本模式之比較	100
第六章 結論與建議	124
6.1 結論	124
6.2 建議	125
參考文獻	126

圖目錄
圖1-1 單一路段(含路口)、雙路段(含路口)..............................................................2
圖1-2 固定式與移動式偵測器時空圖.......................................................................5
圖1-3 整體研究架構圖...............................................................................................6
圖1-4 轉換函數模式示意圖.......................................................................................7
圖1-5 研究流程圖.......................................................................................................8
圖2-1 美國國防部資料融合架構.............................................................................18
圖3-1 模式架構圖.....................................................................................................24
圖3-2 動態系統轉換模式圖.....................................................................................25
圖3-3 包括噪訊之動態系統轉換函數圖.................................................................26
圖3-4 探針車軌跡時空圖.........................................................................................32
圖3-5 軌跡點資訊(速度) ..........................................................................................34
圖3-6 固定式偵測器篩選流程.................................................................................38
圖3-7 探針車篩選過流程.........................................................................................39
圖3-8 固定式之靜態與動態資料庫.........................................................................43
圖3-9 固定式轉換函數演算流程.............................................................................44
圖3-10 移動式動態資料庫.......................................................................................46
圖3-11 移動式轉換函數演算流程...........................................................................47
圖3-12 融合旅行時間演算流程...............................................................................48
圖4-1 路網示意圖.....................................................................................................56
圖4-2 偵測器輸出資料.............................................................................................59
圖4-3 按時階整理之固定式資料.............................................................................59
圖4-4 移動式輸出資料格式.....................................................................................60
圖4-5 個別探針車之軌跡資料.................................................................................60
圖4-6 經計算所得之個別車輛平均速度.................................................................61
圖4-7 個別旅行時間輸出資料.................................................................................61
圖4-8 按時階分之平均路段旅行時間.....................................................................62
圖5-1 路段二速率與密度之靜態關係.....................................................................63
圖5-2 路段二之αt 與βt 之CCF 圖.........................................................................66
圖5-3 路段一之αt 與βt 之CCF 圖.........................................................................66
圖5-4 每0.5 秒之探針車資料..................................................................................70
圖5-5 個別車輛在各時間點之速度.........................................................................70
圖5-6 路段一之αt 與βt 之CCF 圖.........................................................................71
圖5-7 路段二之αt 與βt 之CCF 圖.........................................................................72
圖5-8 模式驗證流程.................................................................................................78
圖5-9 路段一旅行時間比較(按時階)(固定式)........................................................81
圖5-10 路段二旅行時間比較(按時階)(固定式)......................................................83
圖5-11 連續二路段旅行時間比較(按時階)(固定式) ...............................................85
圖5-12 路段一旅行時間比較(按時階)(移動式)......................................................87
圖5-13 路段二旅行時間比較(按時階)(移動式)......................................................89
圖5-14 連續二路段旅行時間比較(按時階)(移動式)...............................................92
圖5-15 路段一旅行時間比較(按時階)(融合)..........................................................94
圖5-16 路段二旅行時間比較(按時階)(融合)..........................................................96
圖5-17 連續二路段旅行時間比較(按時階)(融合)...................................................98
圖5-18 路段一綜合旅行時間比較...........................................................................99
圖5-19 路段二綜合旅行時間比較...........................................................................99
圖5-20 連續二路段綜合旅行時間比較..................................................................100
圖5-21 路段一固定式旅行時間比較(轉換函數與基本模式) ..............................103
圖5-22 路段二固定式旅行時間比較(轉換函數與基本模式) ...............................105
圖5-23 連續二路段固定式旅行時間比較(轉換函數與基本模式) ......................107
圖5-24 路段一移動式旅行時間比較(轉換函數與基本模式) ..............................110
圖5-25 路段二移動式旅行時間比較(轉換函數與基本模式) ..............................112
圖5-26 連續二路段移動式旅行時間比較(轉換函數與基本模式) .......................114
圖5-27 路段一融合旅行時間比較(轉換函數與基本模式) ...................................117
圖5-28 路段二融合旅行時間比較(轉換函數與基本模式) ...................................119
圖5-29 連續二路段融合旅行時間比較(轉換函數與基本模式) ..........................121

表目錄
表1.1 選定路段屬性表................................................................................................3
表1.2 偵測器所蒐集之常見交通資料........................................................................4
表1.3 旅行時間之分類................................................................................................5
表2.1 車輛偵測技術比較表......................................................................................10
表2.2 使用固定式偵測器資料之相關研究（國外）..............................................12
表2.3 國內使用固定式偵測器資料之相關研究.......................................................13
表2.4 使用移動式偵測器資料之相關研究（國外）..............................................15
表2.5 使用移動式偵測器資料之相關研究（國內）...............................................16
表2.6 資料融合技術分類...........................................................................................19
表2.7 各融合技術比較..............................................................................................19
表2.8 情境說明..........................................................................................................22
表3.1 模式預測能力評估準則...............................................................................29
表3.2 個別回報之篩選檢查標籤..............................................................................37
表3.3 融合模式所需參數值(針對號誌化路段).......................................................40
表4.1 PARAMICS 微觀車流模擬軟體子模式說明表.............................................51
表4.2 PARAMICS 模擬網路之規模限制.................................................................51
表4.3 PARAMICS 模擬器駕駛者相關參數說明.....................................................52
表4.4 模擬路段屬性..................................................................................................55
表4.5 PARAMICS 參數預設值之適用性比較表.....................................................57
表4.6 路段流量之參數適合度評估（Mean Reaction Time=0.5）.........................58
表4.7 路段流量之參數適合度評估（Mean Reaction Time=1.0）.........................58
表4.8 都市幹道所需之探針車數量..........................................................................61
表5.1 各路段之靜態速率與密度關係......................................................................63
表5.2 各路段自變數(密度)之模式比較....................................................................64
表5.3 各路段白干擾化之模式..................................................................................65
表5.4 各路段之暫定模式..........................................................................................66
表5.5 各路段之干擾項模式......................................................................................67
表5.6 各路段完整轉換函數模式..............................................................................67
表5.7 校估後之各路段完整轉換函數模式..............................................................68
表5.8 各路段自變數(估算速度)之模式比較...........................................................71
表5.9 各路段白干擾化之模式..................................................................................71
表5.10 各路段之暫定模式........................................................................................72
表5.11 各路段之干擾項模式....................................................................................72
表5.12 各路段完整轉換函數模式............................................................................72
表5.13 校估後之各路段完整轉換函數模式............................................................72
表5.14 各路段在各時階之權重值............................................................................73
表5.15 各路段固定式與移動式旅行時間迴歸係數................................................75
表5.16 各路段之調整因子........................................................................................76
表5.17 固定旅行時間推估模式驗證估算（路段一）............................................79
表5.18 固定旅行時間推估模式驗證估算（路段二）............................................81
表5.19 固定旅行時間推估模式驗證估算(連續兩路段)..........................................83
表5.20 移動式旅行時間推估模式驗證估算（路段一）........................................86
表5.21 移動旅行時間推估模式驗證估算（路段二）............................................88
表5.22 移動旅行時間推估模式驗證估算(連續二路段).........................................90
表5.23 融合旅行時間推估模式驗證估算（路段一）............................................92
表5.24 融合旅行時間推估模式驗證估算（路段二）............................................94
表5.25 融合旅行時間推估模式驗證估算（連續二路段）.....................................96
表5.26 路段一基本模式旅行時間推估(固定式)...................................................101
表5.27 路段二基本模式旅行時間推估(固定式)...................................................103
表5.28 連續二路段基本模式旅行時間推估(固定式)...........................................106
表5.29 路段一基本模式旅行時間推估(移動式)...................................................108
表5.30 路段二基本模式旅行時間推估(移動式)...................................................110
表5.31 連續二路段基本模式旅行時間推估(移動式)...........................................112
表5.32 各路段固定式與移動式旅行時間迴歸係數(基本模式)...........................115
表5.33 路段一基本模式旅行時間推估(融合).......................................................115
表5.34 路段二基本模式旅行時間推估(融合).......................................................117
表5.35 連續二路段基本模式旅行時間推估(融合)...............................................120
表5.36 路段一推估旅行時間比較(轉換函數模式與基本模式)...........................122
表5.37 路段二推估旅行時間比較(轉換函數模式與基本模式)...........................122
表5.38 連續二路段推估旅行時間比較(轉換函數模式與基本模式)...................123

參考文獻
1.Tavana , H., Mahmassani , H.S.,”Estimation and Application of Dynamic Speed-Density Relations by Using Transfer Function Models.”, Transportation Research Record ,pp.47-57., 2000.
2.許雅惠，『以模擬分析法探討轉換函數在預測旅行時間之運用』，淡江大學運輸科學研究所碩士論文，民國93年。
3.交通部，智慧型交通資訊蒐集、處理、傳播與旅行者行為之系列研究－號誌化道路路況資訊偵測方法與省道路段固定式偵測器佈設規劃，民國95年3月。
4.Cesar A. Quiroga, Darcy Bullock, “dies with global positioning and geographic information systems: an integrated methodology”, TR part C.6 p.101-127, 1998
5.交通部運輸研究所，公路行車時間資訊管理系統之規劃研究(1/4)－高速公路部份及模式先期研究，民國93年6月。
6.Oh, J. S.,Jayakrishnan, Recker, W. “Section Travel Time Estimation from Point Detction.”,Presentation at 82th Annual Meeting of Transportataion Research Board, Washington, D.C.,2002.
7.交通部運輸研究所，交通資料偵測與蒐集，民國85 年6 月。
8.Kim, Young Chan “Estimation of Link Travel Time Using Vehicular Detection Devices in TRACS (Traffic Adaptive Control System).”, Presented at the 5th World Congress on Intelligent Transport Systems, October 12-16, 1998, Seoul, Korea.
9.Son, Young Tae，（1998），Estimating Arterial Link Travel Times Using Loop Detector Output. Proceedings of the 5th World Congress on Intelligent Transport Systems
10.Lee, Young-Ihn “Development of a Link Travel Time Prediction Algorithm for Urban Expressway.”, Presented at the 5th World Congress on Intelligent Transport Systems, October 12-16, 1998, Seoul, Korea.
11.Kurokawa, Takeshi “A Study on Travel Time Prediction Method on Inter-City Expressways Using Traffic Capacity at the Bottleneck.”, Presented at the 5th World Congress on Intelligent Transport Systems, October 12-16, 1998, Seoul, Korea.
12.Matsumura, Satoshi “Experimental Verification of Travel Time Prediction Method.”, Presented at the 5th World Congress on Intelligent Transport Systems, October 12-16, 1998, Seoul, Korea.
13.Darryn Paterson “Dynamic Travel Time Estimation on Instrumented Freeways.”, Presented at the 6th World Congress on Intelligent Transport Systems, October, 1999, Toronto, Canada.
14.Hironori Suzuki “Application of a Neural-Kalman-Filter (NKF) Technique for Dynamic Estimation of OD Travel Time and Flow.”, Presented at the 6th World Congress on Intelligent Transport Systems, October, 1999, Toronto, Canada.
15.Diana Jasperse “Real-time Estimation of Travel Time and Queue-Lengths: A Practical Study.”, Presented at the 6th World Congress on Intelligent Transport Systems, October, 1999, Toronto, Canada.
16.Laurence R.Rilett “Estimating Confidence Interval for Freeway Corridor Travel Time Forecasts.”, Presented at the 6th World Congress on Intelligent Transport Systems, October, 1999, Toronto, Canada.
17.Rik van Grol “DACCORD: On-Line Travel Time Estimation/Prediction Results.”, Presented at the 6th World Congress on Intelligent Transport Systems, October, 1999, Toronto, Canada.
18.Fumitaka Kurauchi “An Evaluation on Effect of Travel Time Information from Real-time Origin-Destination Matrices Estimation Model.”, Presented at the 7th World Congress on Intelligent Transport Systems, November 6-9, 2000, Turin, Italy.
19.Peter Bohnke “Automatic Measurement of Travel Time for Motorized Individual Traffic Inside Cities.”, Presented at the 7th World Congress on Intelligent Transport Systems, November 6-9, 2000, Turin, Italy.
20.Giovanni Huisken “Congestion Prediction on Motorways: A Comparative Analysis.” Presented at the 7th World Congress on Intelligent Transport Systems, November 6-9, 2000, Turin, Italy.
21.Carlo Di Taranto “Network Status Estimation and Traffic Prediction in Urban and Sub-urban Areas.”, Presented at the 7th World Congress on Intelligent Transport Systems, November 6-9, 2000, Turin, Italy.
22.張修榕，『高速公路旅行時間之研究』，中央大學研究所碩士論文，民國90年。
23.溫志元，『高速公路進口匝道匯流路段旅行時間研究』，中央大學研究所碩士論文，民國92年。
24.黃裕文，『高速公路施工路段旅行時間預測之研究』，中央大學研究所碩士論文，民國92年。
25.陳彥佑，『應用有限偵測器推估公路車流量之研究』，交通大學運輸科技與管理學系碩士論文，民國94年6月。
26.李鈺雯，都市幹道動態旅行時間推估與交通偵測設施佈設準則之研究，淡江大學運輸科學研究所，民國九十四年。
27.Ghassan Jarjees “Methods for Predicting Bus Travel Times Using a Signpost Position System.”, Presented at the 4th World Congress on Intelligent Transport Systems, October 21-24, 1997, Berlin, Germany.
28.Seung-Won Shin “Randomness of De-noised Time Series Data for Travel Time Forecasting.”, Presented at the 6th World Congress on Intelligent Transport Systems, October, 1999, Toronto, Canada.
29.Laurence R.Rilett “Estimating Confidence Interval for Freeway Corridor Travel Time Forecasts.”, Presented at the 6th World Congress on Intelligent Transport Systems, October, 1999, Toronto, Canada.
30.Dr.Jean-Luc Ygnace “Cellular Phone Positioning and Travel Time Estimates.”, Presented at the 7th World Congress on Intelligent Transport Systems, November 6-9, 2000, Turin, Italy.
31.Tuomo “Effectiveness of Licence Plate Recognition Based Journey Time Monitoring The Demanding Nordic Environment .”, Presented at the 7th World Congress on Intelligent Transport Systems, November 6-9, 2000, Turin, Italy.
32.Masahiko Maruyama “Travel Time Measurement Using the Automated Vehicle Location System.”, Presented at the 7th World Congress on Intelligent Transport Systems, November 6-9, 2000, Turin, Italy.
33.G.Jarjees “Prediction of Vehicle Travel Times Using Ancillary Information Sources.”, Presented at the 7th World Congress on Intelligent Transport Systems, November 6-9, 2000, Turin, Italy.
34.Takayuki Nakata, “Mining Traffic Data from Probe-Car System for Travel Time Prediction”, Industry/Government Track Poster, 2004.
35.吳佳峰，『有GPS資訊提供下之車輛旅行時間預估模式之研究』，交通大學研究所碩士論文，民國90年。
36.林士傑，『高速公路旅行時間預測模式之研究-類神經網路之應用』，成功大學研究所碩士論文，民國90年。
37.張惠汶，『利用公車GPS定位資料推估路段行車速率之研究』，交通大學研究所碩士論文，民國91年。
38.李季森，『應用探測車法預測高速公路旅行時間』，中央大學研究所碩士論文，民國91年。
39.張慶麟，『應用自動車輛辨識預測高速公路路段旅行時間』，中央大學研究所碩士論文，民國92年。
40.Lawrence A. Klein, “Sensor Technologies and Data Requirements for ITS”,2001, Artech House.
41.D. L. Hall and R. J. Linn, “A Taxonomy of Algorithm for Multisensor Data Fusion”, Technical Proceedings of the Joint Service Data Fusion Symposium,1990,Vol. I, DFS-90, pp594-610.
42.Z. Wall, D. J. Dailey “An Algorithm for Predicting the Arrrival Time of Mass Transit Vehicles Using Automatic Vehicle Location Data” ransportation Research Board 78th Annual Meeting January 10-14, 1999 Washington, D.C.
43.Bing Ma, “PARAMETRIC AND NONPARAMETRIC APPROACHES FOR MULTISENSOR DATA FUSION,” A dissertation submitted in partial fulfillment of the requirements for the degree of Doctor of Philosophy (Electrical Engineering:Systems) in The University of Michigan, 2001.
44.Harris, C.J. “Distributed Estimation, Inferencing and Multi-sensor Data Fusion for Real Time Supervisory Control.” Artificial Intelligence in Real-Time Control 1989. Proceedings of the IFAC Workshop. 19-21 Sept. 1989: Shenyang, China (19-24).
45.Ivan, J. N. et al., “Real- Time Data Fusion for Arterial Street Incident Detection Using Neural Networks”, Transportation Research Record, No. 1497, pp. 27-35, 1995.
46.李穎，『類神經網路應用於國道客運班車旅行時間預測模式之研究』，成功大學研究所碩士論文，民國91年。
47.Niehaus, A. and R.F. Stengel. “Probability-based Decision Making for Automated Highway Driving.” VNIS ‘91. Vehicle Navigation & Information Systems Conference Proceedings. 20-23 Oct. 1991: Dearborn, MI. Soc. Automotive Eng.:Warrendale, PA, 1991. Vol. 2: (1125-36).
48.交通部，「智慧型交通資訊蒐集、處理傳播與旅行者行為系列之研究－號誌化道路路況資訊偵測方法與格式訂定」，民國92年11月。
49.交通部，「智慧型交通資訊蒐集、處理、傳播與旅行者行為之系列研究─號誌化道路路況資訊偵測方法與格式訂定(二)」，民國93年11月。
50.Berka, S. , Tian, X., Changes in Parameters of Data Fusion Subcomponent, Memo, Urban Transportation Center, University of Illinois at Chicago, 1994.
51.S. Berka, A. Tarko, N. Rouphail, V. Sisiopiku and D. Lee, Data Fusion Algorithm for ADVANCE Release 1.5, ADVANCE WORKING PAPER SERIES #37, February 1995.
52.薛聖弘，應用PARAMICS微觀交通模擬軟體於都市路網規劃設計之研究，成功大學研究所碩士論文，民國89年。
53.交通部運輸研究所，『先進式微觀車流模擬器-PARAMICS模擬軟體應用台灣地區發展ITS模擬路網之模式校估測試研究』，民國90年9月。
54.Turner, S. M., William L. Eisele, Robert J. Benz, and Douglas J. Holdener “Travel Time Data Collection Handbook”, Federal Highway Administration, U.S DOT Technical Report, No. FHWA-PL-98-035, (1998)
55.蔡百里，資料融合技術應用於旅行時間推估之研究，淡江大學運輸科學研究所，民國95年
56.林茂文，時間數列分析與預測，華泰書局，1992年11月初版。
57.吳柏林，時間數列分析導論，華泰書局，1995年9月初版。
58.葉小蓁，時間數列分析與應用，台北市，1998年初版。