本研究是使用嵌入式系統之方式來開發車輛之自動駕駛系統，主要分成三個部分：一為使用圖像辨識法來辨識車道線建構車道維持系統，透過拍攝車輛前方之道路，輔以灰階化、座標變換等影像處理方式來偵測出車道線的位置，並將標示出的道路線影像座標轉至鳥瞰圖之二維空間座標，以計算出車道的曲率，以此為基礎再利用純追跡法令車輛追蹤目標車道；二是對路徑中的障礙物進行迴避，本研究使用了雷射掃描儀偵測車輛前方是否存在障礙物，以避障追跡法規劃車輛之行進路線，進而得到車輛轉向角度指令；三為利用YOLO v3物件偵測來辨識基本號誌之號誌辨識系統，利用卷積神經網路來辨識所需要感知的物體，並以辨識結果來規劃車輛號誌決策。最後以嵌入式系統的方式，以六分之一實車大小之電動車進行測試與驗證，實際測試路況上設計直線與曲線路段並放置障礙物、紅綠燈、行車號誌，實際測試此研究之路徑規劃及避障能力。

In this research, embedded system is used to develop autonomous driving system for a small electric vehicle. The discussion topic can be divided into three parts. The first part is to establish an onboard lane keeping system for vehicle driving based on the image processing technique. According to the road picture in front of the vehicle, gray scale and Perspective Transformation image processing technique were used to detect the lane line position. Then, the marked lane line image coordinates were transformed to the 2-D space coordinates of the bird’s-eye view for calculating the curvature of the lane. Hence, the vehicle tracking target lane can be extracted by using pure tracking method. The second part is to investigate the obstacles avoidance of the moving path. LiDar was used to detect the obstacles in front of the vehicle. The obstacle avoidance tracking method was used to plan the route of the vehicle and obtain the command of the vehicle steering angle. The third part is to establish the basic traffic sign recognition system based on the image processing technology. The convolutional neural network were employed to identify the objects and traffic sign location. Then, those traffic sign information were used to detect the vehicle driving strategy based on traffic rules. All the developing hardware and software structure were embedded on one-sixth size of electric vehicle for driving test and verification. The obstacles, traffic lights, and traffic signs are placed on the roadside of a testing side with straight and curved road section for evaluating the autonomous driving system performance.

【1】 SAE。SAE Taxonomy and Definitions for Terms Related to Driving Automation Systems for On-Road Motor Vehicles。民109年6月15日，
取自：https://www. sae. org/standards/content/j3016_201806/
【2】 NHTSA。National Highway Traffic Safety Administration Preliminary Statement of Policy Concerning Automated Vehicles。民109年6月15日。
取自：https://one. nhtsa. gov/Research/Crash-Avoidance/Automated-Vehicles
【3】 劉少山、唐傑、吳雙、李立耘。無人駕駛真的來了：第一本從技術面深入的實作書。佳魁數位。台灣。2018
【4】 D. C. Shoup, “Cruising for parking,” in Access Magazine, 1(30), pp. 16-22, April 2007.
【5】 D. J. LeBlanc, G.E. Johnson, P.J.T. Venhovens, G. Gerber, R. DeSonia, R.D. Ervin, Chiu-Feng Lin, A.G. Ulsoy, T.E. Pilutti, "CAPC: an implementation of a road-departure warning system," Proceeding of the 1996 IEEE International Conference on Control Applications IEEE International Conference on Control Applications held together with IEEE International Symposium on Intelligent Contro, Dearborn, MI, USA, 1996, pp. 590-595, doi: 10.1109/CCA.1996.558926.
【6】 Mei Chen, T. Jochem and D. Pomerleau, "AURORA: a vision-based roadway departure warning system," Proceedings 1995 IEEE/RSJ International Conference on Intelligent Robots and Systems. Human Robot Interaction and Cooperative Robots, Pittsburgh, PA, USA, 1995, pp. 243-248 vol.1, doi: 10.1109/IROS.1995.525803.
【7】 K. Guo, E. G. Hoare, D. Jasteh, X. Sheng and M. Gashinova, "Road Edge Recognition Using the Stripe Hough Transform From Millimeter-Wave Radar Images," in IEEE Transactions on Intelligent Transportation Systems, vol. 16, no. 2, pp. 825-833, April 2015, doi: 10.1109/TITS.2014.2342875.

【8】 R. K. Satzoda, S. Sathyanarayana, T. Srikanthan and S. Sathyanarayana, "Hierarchical Additive Hough Transform for Lane Detection," in IEEE Embedded Systems Letters, vol. 2, no. 2, pp. 23-26, June 2010, doi: 10.1109/LES.2010.2051412.
【9】 H. Cheng, B. Jeng, P. Tseng and K. Fan, "Lane Detection With Moving Vehicles in the Traffic Scenes," in IEEE Transactions on Intelligent Transportation Systems, vol. 7, no. 4, pp. 571-582, Dec. 2006, doi: 10.1109/TITS.2006.883940.
【10】 D. Neven, B. D. Brabandere, S. Georgoulis, M. Proesmans and L. V. Gool, "Towards End-to-End Lane Detection: an Instance Segmentation Approach," 2018 IEEE Intelligent Vehicles Symposium (IV), Changshu, China, 2018, pp. 286-291, doi: 10.1109/IVS.2018.8500547.
【11】 R. S. Mamidala, U. Uthkota, M. B. Shankar, A. J. Antony and A. V. Narasimhadhan, "Dynamic Approach for Lane Detection using Google Street View and CNN," TENCON 2019 - 2019 IEEE Region 10 Conference (TENCON), Kochi, India, 2019, pp. 2454-2459, doi: 10.1109/TENCON.2019.8929655.
【12】 Y. Guo, G. Chen, P. Zhao, W. Zhang, J. Miao, J. Wang，Tae Eun Choe “Gen-LaneNet: A Generalized and Scalable Approach for 3D Lane Detection arXiv:2003.10656 [cs.CV]。民109年5月13日，
取自：https://arxiv. org/abs/2003. 10656
【13】 P. E. Hart, N. J. Nilsson and B. Raphael, "A Formal Basis for the Heuristic Determination of Minimum Cost Paths," in IEEE Transactions on Systems Science and Cybernetics, vol. 4, no. 2, pp. 100-107, July 1968, doi: 10.1109/TSSC.1968.300136.
【14】 P. U. Lee, D. C. Ruspini and O. Khatib, "Dynamic simulation of interactive robotic environment," Proceedings of the 1994 IEEE International Conference on Robotics and Automation, San Diego, CA, USA, 1994, pp. 1147-1152 vol.2, doi: 10.1109/ROBOT.1994.351206.
【15】 D. Fox, W. Burgard and S. Thrun, "The dynamic window approach to collision avoidance," in IEEE Robotics & Automation Magazine, vol. 4, no. 1, pp. 23-33, March 1997, doi: 10.1109/100.580977.
【16】 S. C. Diamantas and R. M. Crowder, "Localisation and Mapping Using a Laser Range Finder: A Goal-Seeking Approach," 2009 Fifth International Conference on Autonomic and Autonomous Systems, Valencia, Spain, 2009, pp. 270-276, doi: 10.1109/ICAS.2009.59.
【17】 F. He, Z. Du, X. Liu and Y. Ta, "Laser range finder based moving object tracking and avoidance in dynamic environment," The 2010 IEEE International Conference on Information and Automation, Harbin, China, 2010, pp. 2357-2362, doi: 10.1109/ICINFA.2010.5512068.
【18】 N. A. Rawashdeh and H. T. Jasim, "Multi-sensor input path planning for an autonomous ground vehicle," 2013 9th International Symposium on Mechatronics and its Applications (ISMA), Amman, Jordan, 2013, pp. 1-6, doi: 10.1109/ISMA.2013.6547399.
【19】 R. Girshick, J. Donahue, T. Darrell and J. Malik, "Rich Feature Hierarchies for Accurate Object Detection and Semantic Segmentation," 2014 IEEE Conference on Computer Vision and Pattern Recognition, Columbus, OH, USA,2014, pp. 580-587, doi: 10.1109/CVPR.2014.81.
【20】 R. Girshick, "Fast R-CNN," 2015 IEEE International Conference on Computer Vision (ICCV), Santiago, Chile, 2015, pp. 1440-1448, doi: 10.1109/ICCV.2015.169.
【21】 R. Girshick, J. Donahue, T. Darrell and J. Malik, "Region-Based Convolutional Networks for Accurate Object Detection and Segmentation," in IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. 38, no. 1, pp. 142-158, 1 Jan. 2016, doi: 10.1109/TPAMI.2015.2437384.
【22】 J. Redmon, S. Divvala, R. Girshick, A. Farhadi，You Only Look Once: Unified, Real-Time Object Detection arXiv:1506.02640 [cs.CV]，民109年5月15日，
取自：https://arxiv. org/abs/1506. 02640
【23】 W. Liu, D. Anguelov, D. Erhan, C. Szegedy, S. Reed, C.Y. Fu, A. Berg。SSD: Single Shot MultiBox Detector arXiv:1512.02325 [cs.CV]。民109年6月22日，
取自：https://arxiv. org/abs/1512.02325
【24】 SAE Mobilus。SAE J670 : Vehicle Dynamics Terminology。民109年5月22日，
取自：https://saemobilus. sae. org/content/j670_200801
【25】 祥儀企業。RA-37GM。民109年6月4日，
取自：http://www. shayangye. com/product-inner.aspx?f=s&i=184
【26】 HOKOYO。URG-04LX-UG01。民109年6月4日，
取自：https://www. hokuyo-aut. jp/search/single. php?serial=166
【27】 Arduino。Mega 2560。民109年6月4日，
取自：https://store. arduino. cc/usa/mega-2560-r3
【28】 NVIDIA。Jetson TX2。民109年6月4日，
取自：https://www. nvidia. com/en-us/autonomous-machines/
embedded-systems/jetson-tx2/
【29】 羅技。Camera C525。民109年6月4日，
取自：https://www.logitech.com/zh-tw/product/hd-webcam-c525
【30】 ROS。ROS資訊。民109年6月4日，
取自：https://www.ros.org /
【31】 胡春旭，ROS機器人開發實踐，機械工業出版社，中國，2018
【32】 維基百科。Sobel。民109年6月4日，
取自：https://en. wikipedia. org/wiki/Sobel_operator
【33】 維基百科。HLS。民109年6月4日，
取自：https://en. wikipedia. org/wiki/HSL_and_HSV
【34】 林宗翰，“電腦視覺與應用上課講義”，國立台灣科技大學色彩與照明科技研究所
【35】 維基百科。Image histogram。民109年6月4日，
取自：https://en. wikipedia. org/wiki/Image_histogram

【36】 維基百科。Mahalanobis distance。民109年6月4日，
取自：https://en. wikipedia. org/wiki/Mahalanobis_distance
【37】 C. Szegedy, W. Liu, Y. Jia, P. Sermanet, S. Reed, D. Anguelov, D. Erhan, V. Vanhoucke, A. Rabinovich。Going deeper with convolutions arXiv:1409.4842 [cs.CV]。民109年6月10日，
取自：https://arxiv. org/abs/1409. 4842
【38】 Darknet。Darknet-53。民109年6月4日，
取自：https://pjreddie. com/
【39】 CSDN。yolo structure。民109年6月4日，
取自：https://blog. csdn. net/leviopku/article/details/82660381
【40】 J. Jing, D. Zhuo, H. Zhang, Y. Liang, M. Zheng。Fabric defect detection using the improved YOLOv3 model。民109年5月18日，
取自：https://doi. org/10. 1177/1558925020908268
【41】 CSDN。loss function 。民109年6月10日，
取自：https://blog. csdn. net/weixin_43384257/article/details/100986249
【42】 黃緒哲， “智慧型控制系統上課講義”，國立台灣科技大學機械工程研究所
【43】 L. A. Zadeh, "Outline of a New Approach to the Analysis of Complex Systems and Decision Processes," in IEEE Transactions on Systems, Man, and Cybernetics, vol. SMC-3, no. 1, pp. 28-44, Jan. 1973, doi: 10.1109/TSMC.1973.5408575.
【44】 E. H. Mamdani, "Application of fuzzy algorithms for control of simple dynamic plant," in Proceedings of the Institution of Electrical Engineers, vol. 121, no. 12, pp. 1585-1588, December 1974, doi: 10.1049/piee.1974.0328.
【45】 R. Craig Coulter, “Implementation of the Pure Pursuit Path Tracking Algorithm, “ The Robotics Institute Camegie Mellon University Pittsburgh, Pennsylvania 15213, January 1992

【46】 V. J. Lumelsky, A. A. Stepanov, “Path-planning strategies for a point mobile automaton moving amidst unknown obstacles of arbitrary shape,” Algorithmica 2, 403–430 (1987).
【47】 薛詩翰，小型電動車之可行使區域搜索與定位，碩士論文，國立台灣科技大學機械工程所，台灣，2019