近年來，先進駕駛輔助系統發展迅速，成為車廠積極發展的智慧車輛技術之一。自動駕駛車爲了安全起見，不只需要先進駕駛輔助系統做為前身發展，速度規劃使人乘車感到舒適也是很重要的一環。先前的文獻僅以無障礙物的前提之下做出速度規劃，然而在實際生活中，這樣的場景可說是少之又少。障礙物的出現可能是靜態也可能是動態的，如果未能及時的偵測這些障礙物且立即進行速度規劃的修正，乘客極可能有不舒適的感受，甚至於造成事故的發生。
本論文使用多感測器融合技術來設計及實現一個車輛速度規劃系統。本論文使用以差分全球衛星定位系統、慣性量測儀、光學編碼器等多感測器的融合技術和擴展式卡爾曼濾波器、方位推估法，並加上地圖比對法進行車輛定位。透過車輛定位得知當下車輛位置後，此系統透過雷射測距儀即時偵測前方障礙物，並應用改良之動態速度規劃演算法，以達到及時且快速地速度修正規劃的目的。綜合以上定位及動態速度規劃兩大子系統，本系統能夠動態規劃出適合當下位置的對應速度。經實驗證明，無論有無論靜態或動態障礙物的情形下，本系統能有效率的即時掌握速度並及時反應。實驗結果也顯示出動態調整之車輛速度規劃系統確實比無障礙物之速度規劃系統的誤差更小，更符合實際所行走之速度變化，也更符合實際道路的行駛狀況。

Recently, advanced driver assistance systems (ADAS) have been explosively developing by most of intelligent vehicle manufacturers. Autonomous vehicles need not only the ADAS for safety purpose but also the smooth speed planning systems to give passengers a comfortable riding condition. However, until now, the previous speed planning systems are under the situations that the roads have no obstacles. Nevertheless, in real road conditions, it is rare. Obstacles can be static, such as dropped furniture; or dynamic, such as low speed vehicles or pedestrians. If the speed planning system cannot detect those obstacles in time and calculate a series of smooth speeds in real-time, the passengers will feel uncomfortable and even the vehicles incur disastrous vehicle accidents.
In this paper, we use a multi-sensor fusion technology that fuses a differential global positioning systems (DGPS), an inertial measurement unit (IMU), and an optical encoder (OE) to provide multiple types of information. Furthermore, by utilizing the information, we adopt the extended Kalman filter (EKF) algorithm, the dead reckoning method (DR) and the map matching (MM) method to indicate the current vehicle’s position. By the use of a Laser Range Finder (LRF) to detect the obstacles, the system uses an improved and fast dynamic speed planning algorithm to detect those obstacles in time and calculate a series of smooth speeds in real-time. By cooperating with the positioning subsystem and the speed planning subsystem, the system can plan an appropriate speed for the current position of the vehicle. Experimental results show that, no matter whether there are static or dynamic obstacles or not, the system can effectively and efficiently control the speed of the vehicle in real-time and react in time. Experimental results also show that the error of the dynamic speed planning system is surely smaller than the static speed planning system. The system is also more practical in real situations of general roads.

目 錄
論文審定書 i
致 謝 ii
摘 要 iii
Abstract iv
目 錄 vi
圖 次 viii
表 次 xi
第一章 緒 論 1
1-1 研究動機 1
1-2 文獻回顧 3
1-3 主要貢獻 6
1-4 章節介紹 6
第二章 系統概述 8
2-1 車輛速度規劃系統設計 8
2-2 兩大系統功能 8
2-2-1 車輛定位系統 8
2-2-2 速度規劃系統 9
2-3 系統架構流程 10
第三章 系統設計與實現 12
3-1 實驗平台 12
3-2 車輛定位系統 13
3-2-1 全球定位系統 13
3-2-2 慣性量測單元 15
3-2-3 光學編碼器 16
3-2-4 方位推估法 19
3-2-5 擴展式卡爾曼濾波器 20
3-2-6 方位推估法與擴展式卡爾曼濾波器之整合 27
3-2-7 地理資訊系統 27
3-2-8 地圖比對法 32
3-3 速度規劃系統 34
3-3-1 雷射測距儀 34
3-3-2 速度規劃系統 37
第四章 實驗結果 49
4-1 實驗場景 49
4-2 車輛定位系統 51
4-2-1 EKF驗證結果 51
4-2-2 定位實驗結果 53
4-3 速度規劃系統 55
4-3-1 模擬結果 55
4-3-2速度規劃實驗結果 65
第五章 結論與未來展望 74
5-1 結論 74
5-2 未來展望 74
參考文獻 76
附件A 79

[1]新通訊快報, http://www.infocom.tw/Page_20150331_智慧汽車.htm, 2015.
[2]智庫百科-車聯網, http://wiki.mbalib.com/zh-tw/车联网.
[3]CESA Automotive Electronics, http://cesa-automotive-electronics.blogspot.tw/ 2014_09_01_archive.html.
[4]GSMA, http://www.gsma.com/connectedliving/wp-content/uploads/2012/06/cl_ mauto_eCall_2906121.pdf.
[5]車測中心，“自動駕駛車發展現況與展望”, 2014.
[6]S. Rezaei, and R. Sengupta, “Kalman Filter-Based Integration of DGPS and Vehicle Sensors for Localization”, IEEE Control System Technology, pp. 1080-1088, 2007.
[7]Shun-Hung Chen, Chan Wei Hsu, and Shih Chieh Huang, “Recursive Estimation of Vehicle Position by Using Navigation Sensor Fusion”, 12th International Conference on ITS Telecommunications, pp.532-536, 2012.
[8]李光偉, 吳政欣, 許文賢, “車輛差分定位與車隊管理系統規劃與應用”, 財團法人車輛研究測試中心, 2005.
[9]LI Jie, FU Meng-yin, “Research on Route Planning and Map-matching in Vehicle GPS/Dead-Reckoning/Electronic Map Integrated Navigation System”, IEEE Intelligent Transportation Systems, vol. 2, pp. 1639-1643, 2003.
[10]Steven Liu, “An On-line Reference-Trajectory Generator for Smooth Motion of Impulse-Controlled Industrial Manipulators”, IEEE Advanced Motion Control, pp.365-370, 2002.
[11]Jorge Villagra, Vicente Milanes, Joshue Perez, Jorge Godoy, “Smooth path and speed planning for an automated public transport vehicle”, Robotics and Autonomous Systems, pp.252-265, 2012.
[12]Club Car, Owner’s manual of DS golf Car electric vehicles.
[13]安守中，GPS定位原理及應用，全華圖書，2009。
[14]NOAA, http://www.ngs.noaa.gov/CORS/CorsPP/WA-SlideShow/index.htm.
[15]SPAN-SE,USER MANUAL, http://www.novatel.com/assets/Documents/Manuals/ om-20000124.pdf.
[16]NMEA data, http://www.gpsinformation.org/dale/nmea.htm.
[17]Xsens：http://www.xsens.com.
[18]MyTech, http://cht.nahua.com.tw/encoder/.
[19]元智大學-最佳化設計實驗室, http://designer.mech.yzu.edu.tw/article/articles /thesis/Master_90TsaiChungChen/05第五章第二代遠端網路代理.htm.
[20]郭孝立，整合GPS與方位推估法在個人定位之應用，碩士論文，國立台灣科技大學機械工程所，2000。
[21]R.E. Kalman, “A New Approach to Linear Filtering and Prediction Problems”, Research Institute for Advanced Study, Baltimore, Md., 1960.
[22]PARALLAX PROPELLER PROTO BOARD USB, http://www.adrirobot.it/ parallax/propeller/parallax_propeller_protoboard.htm.
[23]林岳霆，結合定位技術與路口代理人之車輛安全防撞系統，碩士論文，國立中山大學機械與機電工程研究所，2010。
[24]林佳慧，擴展行卡爾曼濾波器於車輛定位追蹤之應用，碩士論文，國立臺灣海洋大學電機工程研究所，2009。
[25]行政院主計總處，“何謂地理資訊系統”。http://www.dgbas.gov.tw/ ct.asp?xItem=11385&ctNode=2391.
[26]Vincenty’s formula, http://en.wikipedia.org/wiki/Vincenty's_formulae#External _links.
[27]Converting UTM to Latitude and Longitude (Or Vice Versa), http://www.uwgb. edu/dutchs/UsefulData/UTMFormulas.HTM.
[28]TWD97轉經緯度WGS84經緯度WGS84轉TWD97, http:// wangshifuola.blogspot.tw/2010/08/twd97wgs84-wgs84twd97.html?showComment =1416201723752#c4712257572197205629.
[29]Vector Product, http://www.csie.ntnu.edu.tw/~u91029/Point.html.
[30]SICK, Technical description of LMS-291 laser measurement systems.
[31]雷射測量系統SICK LMS-291, http://es-tech.blogspot.tw/2008/02/sick-lms-291.html.
[32]楊瓊茹、洪萬生，極座標(Polar Coordinate)，高瞻自然科學教學資源平台，科技部。
[33]車輛測試研究中心，油門與煞車踏板的控制。 http://www.artc.org.tw/chinese/03_service/03_02detail.aspx?pid=2226&nPage=2&syear=0&skind1=0&skind2=0&skeyword=煞車.
[34]車輛測試研究中心，“乘車舒適性有標準嗎?-淺談國際標準ISO2631”。 http://www.artc.org.tw/chinese/03_service/03_02detail.aspx?pid=58&nPage=1&syear=0&skind1=0&skind2=0&skeyword=舒適.
[35]Ford, http://www.ford.com.tw/techsection.
[36]Lexus, http://www.lexus.com.tw/app/lexusplus/.
[37]Andy Birnie, Iain Davidson，“實現V2V/V2I通訊，DSRC提高聯網器車安全”，新電子。http://www.mem.com.tw/article_content.asp?sn=1302070003
[38]自由時報，“打破汽車定義!賓士無人概念車F015 Luxurt in Motion”。http://auto.ltn.com.tw/news/1523/3.