隨著CAN FD匯流排的提出，車輛電子的研究方向朝向新一代匯流排發展。但發展至今，大部分對CAN FD 匯流排的研究都偏向比較CAN匯流排和CAN FD匯流排的差異性，並未對CAN FD匯流排有深入的探討與分析。
有鑑於此，本論文以使用CAN FD匯流排為基礎，開發出具備車輛主動安全(ADAS)的ECU節點，分別為Embedded Throttle-CAN FD ECU Node與Embedded Steer-CAN FD ECU Node，節點使用LPC1768撰寫USB Host讀取USB賽車方向盤組的資料，模擬汽車的油門與方向盤。此外，也設計Embedded Dashboard-CAN Node與Embedded Simulated Car-CAN Node驗證CAN FD匯流排對於CAN匯流排的兼容性，形成一完整的CAN/CAN FD車輛混合網路。
為了CAN FD匯流排的發展性，本論文採用機器學習方法-RM排程法，並使用邏輯回歸法對優先權分配方式分類，找出CAN FD匯流排在不同資料量及優先權的情況下透過改變優先權的方式來改善傳輸的方法。本論文先隨機產生100~1000筆資料依照優先權與資料長度使用邏輯回歸法分為10個類別，且不同類別給予不同CAN FD ID決定優先權。之後，再將資料打入CAN FD分析儀內進行測試。在改善CAN FD匯流排傳輸的部分，本論文以資料遺失率作為依據，使用PCAN-USB Pro FD(PEAK System)測試。而所有的CAN FD節點的資料，都應用本論文研發的USB 3.1 Gen1-CAN FD Analyzer進行接收，並透過自行撰寫的分析儀程式進行資料驗證的部分。
最後，經過上述的驗證與測試後，證明透過改變不同週期與資料量的CAN FD Bus的優先權，在資料遺失的比率上平均減少了2.13%。

The research direction of automotive electronics has evolved with the invention of a controller area network with flexible data (CAN FD) bus. This study developed an engine control unit (ECU) node with an advanced driver assistance systems (ADAS) based on the CAN FD bus, which combines an embedded Throttle-CAN FD ECU node and an embedded Steer-CAN FD ECU node. The nodes used the LPC1768 to control the USB host to read the information of the USB steering wheel set, simulating the throttle and steering wheel of a car. Additionally, a Dashboard CAN node and the Simulated Car CAN node were designed to verify the compatibility of the CAN FD bus and the CAN bus to form a complete a CAN/CAN FD hybrid vehicle network.
To develop the CAN FD bus, this study used rate monotonic scheduling to integrate machine learning. First, logistic regression was used to classify data priorities and identify the CAN FD bus from different data quantities and priorities. The method improved the data transmission by changing the priority. This study first used a self-developed analyzer to generate 100–1000 data and divide them into ten categories according to priority and length; the categories allowed the CAN FD ID to determine the priority. Subsequently, the data were tested on the analyzer of the CAN FD bus. To improve CAN FD bus transmission, this study used the PCAN-USB Pro FD (PEAK System) to test the data loss rate. All CAN FD node data were processed using the self-developed USB 3.1 Gen1-CAN FD Analyzer, and data were verified using a self-developed analyzer program.
After verification and testing, this study proved that by changing the priority of the CAN FD bus of different periods and data volumes, the ratio of data loss was reduced by 2.13%.

摘要..........................................................i
Abstract..........................................................ii
誌謝..........................................................iii
目錄..........................................................iv
表目錄..........................................................vi
圖目錄..........................................................vii
第一章 緒論..........................................................1
1.1 研究背景..........................................................1
1.2 動機與目的.......................................................... 2
1.3 國內外之相關研究情況................................................. 3
1.4 論文章節概述........................................................ 3
第二章 系統原理......................................................... 4
2.1 機器學習介紹........................................................ 4
2.1.1 邏輯回歸法簡介.................................................... 4
2.1.2 邏輯回歸法應用.................................................... 5
2.2 車載通匯流排協定介紹................................................. 6
2.2.1 CAN簡介......................................................... 6
2.2.2 CAN通訊協定之模型架構.............................................. 7
2.2.3 CAN通訊協定之通訊格式.............................................. 8
2.2.4 CAN FD簡介....................................................... 9
2.2.5 CAN FD通訊協定之通訊格式........................................... 9
2.2.6 CAN FD實際效能.................................................... 11
2.3 USB通訊協定......................................................... 12
2.3.1 USB介面架構....................................................... 12
2.3.2 USB介面通訊概念................................................... 14
2.3.3 USB介面傳輸類型................................................... 15
2.3.4 USB3.1 Gen1介紹.................................................. 19
2.4 SPI通訊匯流排....................................................... 21
2.4.1 SPI主從介面簡介................................................... 21
2.4.2 SPI的通訊協定..................................................... 22
第三章 系統設計.........................................................24
3.1 系統架構.........................................................24
3.2 Embedded CAN FD/CAN Gateway (CAN FD Node2)設計...................25
3.3 CAN FD節點設計.................................................. 26
3.3.1 Embedded Throttle-CAN FD ECU Node(CAN FD Node3)設計..............26
3.3.2 Embedded Steer-CAN FD ECU Node(CAN FD Node4)設計.................27
3.4 CAN節點設計......................................................28
3.4.1 Embedded Dashboard-CAN Node(CAN Node1)設計.......................28
3.4.2 Embedded Simulated Car-CAN Node (CAN Node2)設計..................31
3.5 USB 3.1 Gen1-CAN FD Analyzer(CAN FD Node1)設計...................32
3.6 邏輯回歸排程設計................................................ 37
3.6.1 產生測試的CAN FD資料..............................................37
3.6.2 邏輯回歸分類......................................................39
第四章 系統測試..........................................................45
4.1 CAN FD通訊驗證....................................................46
4.2 CAN 通訊驗證.....................................................49
4.3 USB Logitech G27 賽車方向盤資料驗證...............................50
4.4 SPI 主從介面通訊驗證...............................................52
4.5 USB 3.1 Gen1- CAN FD Analyzer(CAN FD Node1)測試與驗證..............53
4.6 Embedded CAN FD/CAN Gateway (CAN FD Node2)測試....................57
4.7 CAN FD節點測試...................................................58
4.7.1 Embedded Throttle-CAN FD ECU Node(CAN FD Node3)測試.............58
4.7.2 嵌入式Steer-CAN FD ECU節點測試...................................59
4.8 CAN節點測試......................................................62
4.8.1 Embedded Dashboard-CAN Node(CAN Node1)測試.......................62
4.8.2 Embedded Simulated Car-CAN Node (CAN Node2)測試...................64
4.9 邏輯回歸排程驗證...................................................65
4.9.1 邏輯回歸設計驗證...................................................65
4.9.2 CAN FD邏輯回歸排程測試.............................................71
第五章 結論與未來展望.....................................................76
5.1 結論..............................................................76
5.2 未來展望..........................................................76
參考文獻..................................................................77
Abstract.................................................................79

[1]MoneyDJ理財網-車用ADAS系統http://newjust.masterlink.com.tw/HotProduct/HTML/Basic.xdjhtm?A=PA107-1.HTML
[2]M. AI Hayani, "Modeling Bus Load on CAN".
[3]J. M. Flores-Arias, M. Ortiz-Lopez, F. J. Quiles-Latorre, V. Pallares and A. Chen, "Complete hardware and software bench for the CAN bus", 2016 IEEE International Conference on Consumer Electronics (ICCE), pp. 211-212, 2016.
[4]T. Lindenkreuz, "CAN FD – CAN with Flexible Data-Rate", Vector Kongress, 2012.
[5]Robert Bosch GmbH, "CAN with Flexible Data-Rate", white paper version 1.1, Germany, August 2011.
[6]Robert Bosch GmbH, "CAN with Flexible Data-Rate", specification version 1.0, Germany, April 2012.
[7]Machine learning-Wiki https://en.wikipedia.org/wiki/Machine_learning
[8]"First-hand information on CAN FD", CAN Newsletter Online, December 2012.
[9]T. Lindenkreuz, "CAN FD – CAN with Flexible Data-Rate", Vector Kongress, 2012.
[10]G. Cena, I.C. Bertolotti, Tingting Hu, A. Valenzano, 2014, "Improving compatibility between CAN FD and legacy CAN devices", 2015 IEEE 1st International Forum on Research and Technologies for Society and Industry Leveraging a better tomorrow (RTSI), 419-426, Sept.
[11]Tien-Hai Nguyen, Bo-Mu Cheon, Jae-Wook Jeon, 2014, "CAN FD performance analysis for ECU re-programming using the CANoe", 2014 The 18th IEEE International Symposium on Consumer Electronics (ISCE 2014), 1-4, June.
[12]Bo-Mu Cheon and Jae-Wook Jeon, 2013, "The CAN FD network performance analysis using the CANoe", 2013 44th International Symposium on Robotics (ISR), 1-5, Oct.
[13]U.D. Bordoloi and S. Samii, 2014, "The Frame Packing Problem for CAN-FD", 2014 IEEE Real-Time Systems Symposium (RTSS), 284-293, Dec.
[14]Meiqin Mao, Yangyang Wang, You Yue,"Multi-time scale forecast for schedulable capacity of EVs based on big data and machine learning", IEEE Energy Conversion Congress and Exposition (ECCE), 2017
[15]Le Luong Vy, Li-Ping Tung, Bao-Shuh Paul Lin, "Big data and machine learning driven handover management and forecasting", IEEE Conference on Standards for Communications and Networking (CSCN), 2017
[16]Fuchun Sun, Guang-Bin Huang, Q. M. Jonathan Wu, "Efficient and Rapid Machine Learning Algorithms for Big Data and Dynamic Varying Systems" , IEEE Transactions on Systems, Man, and Cybernetics: Systems (Volume: 47, Issue: 10), Oct. 2017
[17]Jung Woo Shin, Jung Hwan Oh, Sang Muk Lee, "CAN FD controller for in-vehicle system", 2016 IEEE International SoC Design Conference (ISOCC),Oct. 2016
[18]Hyun Su An, Jae Wook Jeon, "Analysis of CAN FD to CAN message routing method for CAN FD and CAN gateway", 2017 17th International Conference on Control, Automation and Systems (ICCAS),Oct.2017
[19]Shan Ding, Ruifeng Huang, Ryo Kurachi, "A Genetic Algorithm for Minimizing Bandwidth Utilization by Packing CAN-FD Frame", 2016 13th International Conference on Embedded Software and Systems (ICESS),Aug. 2016
[20]Machine Learning in Action,Peter Harrington,April 2012,ISBN 9781617290183
[21]梯度下降法,WIKI百科
[22]邏輯回歸Logistic Regression https://read01.com/zkNKnm.html
[23]許永和，2011，Intelligence 3.0+UBIKIT-6612 智慧車載資通訊實驗系統，台灣優奎士有限公司，高雄市。
[24]CAN with Flexible Data-Rate Specification Version 1.0,BOSCH,2012
[25]CAN FD Flexible Data Rate CAN,Vector
[26]Universal Serial Bus Specification，2000，Specification Revision 2.0.
[27]許永和，2006，微處理機與USB主從介面之設計與應用，儒林圖書有限公司，台北市。
[28]許永和，2009，USB規格與理論剖析 含無限USB與USB 3.0，儒林圖書有限公司，台北市。
[29]LPC546xx User manual
[30]許永和，2011，介面設計與實習 PSOC與感測器實務應用，全華圖書，新北市。
[31]HY32D-LCD，Datasheet.
[32]EZ-USB FX3 SuperSpeed USB 3.0 peripheral controller
http://www.cypress.com/products/ez-usb-fx3-superspeed-usb-30-peripheral-controller
[33]Ricardo De Andrade,Kleber N. Hodel,João Francisco Justo, "Analytical and Experimental Performance Evaluations of CAN-FD Bus", IEEE Access( Volume: 6 ),April 2018.