研究生: |
薛詩瀚 SHIH-HAN HSUEH |
---|---|
論文名稱: |
小型電動車之可行駛區域搜索與定位 Drivable-Region Detection and Positioning of Small Electric Vehicle |
指導教授: |
黃緒哲
Shiuh-Jer Huang |
口試委員: |
陳亮光
Liang-Kuang Chen 郭重顯 Chung-Hsien Kuo |
學位類別: |
碩士 Master |
系所名稱: |
工程學院 - 機械工程系 Department of Mechanical Engineering |
論文出版年: | 2019 |
畢業學年度: | 107 |
語文別: | 中文 |
論文頁數: | 96 |
中文關鍵詞: | 自動駕駛 、車輛定位 、語義分割 、路徑追隨 |
外文關鍵詞: | autopilot, vehicle positioning, semantic segmentation, lane keeping |
相關次數: | 點閱:290 下載:15 |
分享至: |
查詢本校圖書館目錄 查詢臺灣博碩士論文知識加值系統 勘誤回報 |
本研究主要分為兩個部分:一為建立GPS與INS整合之車輛定位系統,並使用高精度之RTK-GPS來得到公分等級精度之定位;第二部分為利用光學攝影鏡頭得到之影像輸入建立行車時之道路維持系統。此部分使用語義分割來辨別車輛前方場景中視為可行道路的區塊,再透過座標之空間轉換、影像模糊化、二值化來得到道路鳥瞰視角的形狀,並以此形狀做骨架提取以作為可行進路線之參考。結合了上述提及的兩個部分,本研究之系統硬體架構為一小型電動車搭載FPGA嵌入式開發板作為控制器,以NVIDIA Jetson TX2嵌入式開發套件作為影像處理運算與整車的決策單元。並開發一Android平台APP使用者介面可供使用者對車輛進行遠端操控命令,以實現本研究車輛點到點自動導航功能。而本研究方法之一大優點為:在沒有特別設計避障策略的情況下,環境單純的情況下,僅靠語義分割得到的結果便能達到所規劃之行駛路徑會繞過障礙物所在位置。另一個優點為本研究使用Google Map API作為地圖資訊,只要在Google地圖有支援的區域,便能得到並使用API規劃的行駛路徑,當變換不同使用區域的情況下,使用者無須再自行建立地圖資訊供車輛定位。最後本研究設計實驗及驗證此方法在狹小之道路中行駛並維持在車道中的能力。
This research can be divided into two main parts. One is to establish a GPS and INS vehicle positioning system. The high-precision RTK-GPS is used to obtain the centimeter-level positioning. The second part is to establish a lane keeping system for vehicle driving based on the image input of camera. Semantic segmentation technique is used to recognize the drivable-region in front of the vehicle. The bird's-eye view which is obtained by homography can let vehicle know the shape of road. Then skeleton extraction from the shape of road is used to be the reference path. The FPGA is used as the hardware of controller, and electric vehicle integrated with a NVIDIA Jetson TX2 embedded development kit as the image processing unit and the driving decision unit. A user interface for Android app is developed for the user to remote the vehicle for achieving the point-to-point navigation. One of the advantages of this research is that in the absence of a specially designed obstacle avoidance strategy, the result of semantic segmentation can achieve the planned driving path bypass the obstacle in a simple environment. Another advantage is that the Google Maps API is used as map information. As long as the area supported by Google Maps, the planned route can be obtained and be used. When the using area are changed, the user doesn’t need to create a map again to provide information for vehicle. The experiment verifies the ability of this method for traveling in the riverside park road and maintaining the electric vehicle in the small lane.
【1】 薛舜益(2018年3月1日)。 NUTONOMY 新加坡無人計程車上路服務嚕。財團法人資訊工業策進會 數位服務創新研究所。2019年7月,取自:
https://www.find.org.tw/index/wind/browse/c66382a572cdb878c3913f2d2fd0be96/
【2】 K. Jo, K. Chu and M. Sunwoo, "Interacting Multiple Model Filter-Based Sensor Fusion of GPS With In-Vehicle Sensors for Real-Time Vehicle Positioning," IEEE Transactions on Intelligent Transportation Systems, vol. 13, no. 1, pp. 329-343, March 2012.
【3】 J. Fang and X. Gong, "Predictive Iterated Kalman Filter for INS/GPS Integration and Its Application to SAR Motion Compensation," in IEEE Transactions on Instrumentation and Measurement, vol. 59, no. 4, pp. 909-915, April 2010.
【4】 A. Y. Hata, F. S. Osorio and D. F. Wolf, "Robust curb detection and vehicle localization in urban environments,"2014 IEEE Intelligent Vehicles Symposium Proceedings, Dearborn, MI, 2014, pp. 1257-1262.
【5】 M. Schreiber, H. Königshof, A. Hellmund and C. Stiller, "Vehicle localization with tightly coupled GNSS and visual odometry,"2016 IEEE Intelligent Vehicles Symposium (IV), Gothenburg, 2016, pp. 858-863.
【6】 Z. Zhang, H. Wang and W. Chen, "A real-time visual-inertial mapping and localization method by fusing unstable GPS," 2018 13th World Congress on Intelligent Control and Automation (WCICA), Changsha, China, 2018, pp. 1397-1402.
【7】 J. Canny, "A Computational Approach to Edge Detection," in IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. PAMI-8, no. 6, pp. 679-698, Nov. 1986.
【8】 J. Sona, H. Yooa, S. Kimb and K.Sohn, "Real-time illumination invariant lane detection for lane departure warning system," Expert Systems with Applications, vol. 42, no. 4, March 2015, pp. 1816-1824
【9】 J. Li, X. Mei, D. V. Prokhorov, D. Tao, “Deep Neural Network for Structural Prediction and Lane Detection in Traffic Scene”. IEEE Trans. Neural Netw. Learning Syst., vol. 28, no.3, pp. 690-703, 2017.
【10】 V. Badrinarayanan, A. Kendall and R. Cipolla, "SegNet: A Deep Convolutional Encoder-Decoder Architecture for Image Segmentation," in IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. 39, no. 12, pp. 2481-2495, 1 Dec. 2017.
【11】 Paszke A., Chaurasia A., Kim S. and Culurciello E. "Enet: A deep neural network architecture for real-time semantic segmentation."arXiv preprint arXiv:1606.02147. (2016)
【12】 Zhao, H., Qi, X., Shen, X., Shi, J., & Jia, J. (2018). " Icnet for real-time semantic segmentation on high-resolution images. " In Proceedings of the European Conference on Computer Vision (ECCV),pp. 405-420.
【13】 Yu, C., Wang, J., Peng, C., Gao, C., Yu, G., and Sang, N. (2018). Bisenet: Bilateral segmentation network for real-time semantic segmentation. In Proceedings of the European Conference on Computer Vision (ECCV)(pp. 325-341).
【14】 S. Lee, J.-S. Kim, J. S. Yoon, S. Shin, O. Bailo, N. Kim, T.-H. Lee, H. S. Hong, S.-H. Han, I. S. Kweon. (2017). " Vpgnet: Vanishing point guided network for lane and road marking detection and recognition. " In Proceedings of the IEEE International Conference on Computer Vision (pp. 1947-1955).
【15】 W. Zhou, S. Worrall, A. Zyner and E. Nebot, "Automated Process for Incorporating Drivable Path into Real-Time Semantic Segmentation," 2018 IEEE International Conference on Robotics and Automation (ICRA), Brisbane, QLD, 2018, pp. 1-6.
【16】 Thomas D. Gillespie, Fundamentals of vehicle dynamics, Society of Automotive Engineers, Inc.400 Commonwealth Drive Warrendale, PA.
【17】 ROS.org | Powering the world's robots : https://www.ros.org/
【18】 冨永 貴樹,RTK-GPS 測位計算アルゴリズム,東京海洋大学,2003
【19】 Firebase Realtime Database :
https://firebase.google.com/products/realtime-database/
【20】 Cityscapes Dataset : https://www.cityscapes-dataset.com/
【21】 KITTI Dataset : http://www.cvlibs.net/datasets/kitti/
【22】 Hitachi-Automotive-And-Industry-Lab, Semantic Segmentation Editor : https://github.com/Hitachi-Automotive-And-Industry-Lab/semantic-segmentation-editor
【23】 Morphological skeleton From Wikipedia : https://en.wikipedia.org/wiki/Morphological_skeleton#Discrete_images
【24】 R. Craig Coulter, “Implementation of the Pure Pursuit Path Tracking Algorithm”, The Robotics Institute Camegie Mellon University Pittsburgh, Pennsylvania 15213, January 1992.
【25】 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
【26】 Shiuh-Jer Huang, Intelligent Control System, Mechanical Engineering Department National Taiwan University of Science and Technology .
【27】 楊宗賢,輪毂馬達電動車之電子差速與速度控制,碩士論文,國立台灣科技大學機械工程所,台灣,2012。
【28】 蔡欽章,基於視覺回授之主動式駕駛輔助系統,碩士論文,國立台灣科技大學機械工程所,台灣,2016。
【29】 林福揚,四輪驅動電動車之視覺導引路徑追蹤控制,碩士論文,國立台灣科技大學機械工程所,台灣,2017
【30】 李聖復,自主電動車雛形機電設計及實現,碩士論文,國立台灣科技大學電機工程所,台灣,2018
【31】 姜嘉瑞,GPS與INS整合之三維定位系統研究,碩士論文,國立臺灣工業技術學院機械工程技術研究所,台灣,1997
【32】 王通溫、曾柏凱、洪銘鴻,自主駕駛系統路徑規劃之技術研究,十八屆車輛工程學術研討會,屏東科技大學車輛工程系,台灣,2013
【33】 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