請用此 Handle URI 來引用此文件:
http://tdr.lib.ntu.edu.tw/jspui/handle/123456789/18954
完整後設資料紀錄
DC 欄位 | 值 | 語言 |
---|---|---|
dc.contributor.advisor | 王立昇(Li-Sheng Wang) | |
dc.contributor.author | Li-Heng Gao | en |
dc.contributor.author | 高立恆 | zh_TW |
dc.date.accessioned | 2021-06-08T01:40:39Z | - |
dc.date.copyright | 2021-02-20 | |
dc.date.issued | 2021 | |
dc.date.submitted | 2021-02-08 | |
dc.identifier.citation | [1]Y. Kanayama, Y. Kimura, F. Miyazaki, and T. Noguchi, 'A Stable Tracking Control Method for an Autonomous Mobile Robot', IEEE International Conference on Robotics and Automation, 1990. [2]M. Elbanhawi, M. Simic, and R. N. Jazar, 'Continuous Path Smoothing for Car-Like Robots Using B-Spline Curves', Journal of Intelligent Robotic Systems, vol. 51, Oct, 2015. [3]J. Park, Y. Kim, 'Collision Avoidance for Quadrotor Using Stereo Vision Depth Maps', IEEE Transactions on Aerospace and Electronic Systems, 2015. [4]J. F. Henriques, R. Caseiro, P. Martins, and J. Batista, 'High-Speed Tracking with Kernelized Correlation Filters', IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. 37, 01, August 2014. [5]M. Vidyasagar, 'Nonlinear Systems Analysis', Prentice-Hall Inc. Englewood Cliffs, N.J., 1978. [6]R. C. Coulter, 'Implementation of the Pure Pursuit Path Tracking Algorithm', Carnegie Mellon University technical report, January 1992. [7]Juan Rada-Vilela, 'The FuzzyLite Libraries for Fuzzy Logic Control', www.fuzzylite.com, 2018. [8]J. Redmon and A. Farhadi, 'YOLOv3: An Incremental Improvement', arXiv.org, 2018. [9]K. M. Passino and S. Yurkovich, Fuzzy Control, Addison Wesley Longman Inc, 522, 1998. [10]https://cdn.stereolabs.com/assets/images/zed-2/zed-2-front.jpg [11]https://cdn.stereolabs.com/assets/datasheets/zed2-camera-datasheet.pdf [12]https://www.axis.com/files/datasheet/ds_m3006v_1485632_en_1607.pdf [13]https://www.msi.com//Laptop/GE70-2PE-Apache-Pro/Specification [14]http://noobeed.com/nb_coord_system.htm [15]https://forum.processing.org/two/discussion/22378/short-term-project-work-paid-opencv-rasterisation-arrays-motor-control-arduino [16]https://e2eml.school/convert_rgb_to_grayscale.html [17]https://clickitupanotch.com/lens-distortion/ [18]https://towardsdatascience.com/lines-detection-with-hough-transform-84020b3b1549?gi=2a7e14b97fcd [19]https://www.stereolabs.com/zed-2/ [20]https://pjreddie.com/media/image/yologo_2.png [21]https://towardsdatascience.com/yolov1-you-only-look-once-object-detection-e1f3ffec8a89 [22]https://www.stereolabs.com/docs/object-detection/images/bbox.png [23]https://en.wikipedia.org/wiki/B-spline [24]https://en.wikipedia.org/wiki/Hough_transform | |
dc.identifier.uri | http://tdr.lib.ntu.edu.tw/jspui/handle/123456789/18954 | - |
dc.description.abstract | 在本論文中,我們利用視覺系統的輔助,建構了一個在室內環境中運動的無人自走車即時路徑追蹤與障礙物迴避策略。採用的視覺系統包括具俯視功能的網路攝影機以及ZED 2立體影像攝影機,分別捕捉標有特殊圖樣之載具、障礙物以及路人障礙物,在室內實驗環境中獲得到無人自走車在空間中的座標位置與姿態資訊及周圍環境障礙物訊息。 有了載具與周圍環境的資訊以後,我們利用帶有約束條件的B-spline curve設計一個滿足載具最小迴轉半徑的路徑,並且分別使用比例控制法與模糊控制法追蹤這個設計好的路徑。 經由實驗結果證明,本論文發展的方法確實可以實現我們的任務目標。 | zh_TW |
dc.description.abstract | In this work, we develop a real-time path-following and collision avoidance methodology for an autonomous vehicle in the indoor environment, assisted by vision system. In order to acquire the spatial coordinates and attitude of the vehicle in the confined experimental region, we design a vision system that allows us to obtain the vehicle's location information and surrounding obstacles. The vision system uses two types of camera, the top-view webcam and vehicle-mounted stereo camera ZED 2 to discern the pattern-specified obstacle and the human obstacle, respectively. Once we have the vehicle and environmental information, we utilize B-spline curve with curvature constraints to design the path which fits vehicle's minimal steering radius, and use two types of control method, Proportional Control and Fuzzy Control, to track the path. From the experimental results, the developed methodology can fulfill the objectives of the mission. | en |
dc.description.provenance | Made available in DSpace on 2021-06-08T01:40:39Z (GMT). No. of bitstreams: 1 U0001-0802202116012600.pdf: 6820667 bytes, checksum: f8857d3d30115a437e91ccd250995b62 (MD5) Previous issue date: 2021 | en |
dc.description.tableofcontents | 口試委員會審定書 # 致謝 i 摘要 ii ABSTRACT iii CONTENTS iv LIST OF FIGURES vii LIST OF TABLES xii Chapter 1 Introduction 1 1.1 Overview 1 1.2 Literature Review 1 1.2.1 A Stable Tracking Control Method for An Autonomous Mobile Robot 1 1.2.2 Continuous Path Smoothing for Car-Like Robots Using B-Spline Curves 2 1.2.3 Collision Avoidance for Quadrotor Using Stereo Vision Depth Maps 2 1.3 The Research Content and Results 2 1.4 The Architecture of The Thesis 3 Chapter 2 System Configuration 4 2.1 Hardware 4 2.1.1 Stereolabs Inc. ZED 2 Stereo Camera 4 2.1.2 AXIS M3006-V Network Camera 5 2.1.3 MSI GE70 2PE Apache Pro 6 2.1.4 INNOVATI Customized Unmanned Autonomous Vehicle Module 6 2.2 Software 7 2.3 System Configuration 7 Chapter 3 Image Processing 9 3.1 The Concept of Digital Image 9 3.2 Image Distortion Calibration 11 Chapter 4 Vehicle Tracking and Obstacle Detection 13 4.1 Image Feature Extraction and Tracking with KCF Algorithm 13 4.2 Vehicle's Attitude Determination 14 4.2.1 Hough Transform 14 4.3 Obstacle Detection from Webcam 16 4.4 Stereo Vision 17 4.4.1 Disparity Map 17 4.4.2 You-Only-Look-Once (YOLO) AI Algorithm 19 4.4.3 Collision Cone 20 Chapter 5 Controller Design and Path-planning 22 5.1 Kinematic Model of Autonomous Vehicle 22 5.2 Path-following Control 25 5.3 B-Spline Curve and Target Waypoint Selection 27 5.3.1 Path Design with B-Spline Curve 28 5.3.2 B-Spline Curve with Curvature Constraints 30 5.3.3 Scanning Segment Strategy 32 5.4 Fuzzy Control 33 5.4.1 Fuzzy Logic Controller 34 5.4.2 Membership Function 34 5.4.3 Fuzzy Rule Base 39 5.4.4 Fuzzy Inference 41 5.4.5 Fuzzy Tracking Control 42 5.5 Path-planning 42 5.5.1 Path-planning for the obstacle captured from the webcam 42 5.5.2 Path-planning for the obstacle captured from Stereo Camera ZED 2 43 Chapter 6 Experiment 44 6.1 Experiment Architecture 44 6.2 Vehicle's Path-following Experiment 44 6.2.1 P Control and Fuzzy Control Experiment 44 6.2.2 Data Analysis 46 6.3 Vehicle's Real-Time Obstacle Avoidance Experiment 49 6.3.1 Human Obstacle on the Path (Right Side) 50 6.3.2 Human Obstacle on the Path (Left Side) 52 6.3.3 Only Human Obstacle on the Path (Left Side) 54 6.3.4 Only Human Obstacle on the Path (Right Side) 56 6.3.5 Human and Pattern Obstacles on the Path (Left Side) 58 6.3.6 Human and Pattern Obstacles on the Path (Right Side) 60 6.3.7 Only Pattern Obstacle on the Path 62 6.3.8 Path-Following and Collision Avoidance with sharp turn 63 Chapter 7 Conclusion and Future Work 65 REFERENCE 66 | |
dc.language.iso | en | |
dc.title | 視覺系統輔助自走車路徑追蹤與障礙物迴避之方法研究 | zh_TW |
dc.title | Research on the Methodology of Vision System-Assisted Path-following and Collision Avoidance for Autonomous Vehicle | en |
dc.type | Thesis | |
dc.date.schoolyear | 109-1 | |
dc.description.degree | 碩士 | |
dc.contributor.oralexamcommittee | 張帆人(Fan-Ren Chang),卓大靖(Dah-Jing Jwo),王和盛(He-Sheng Wang) | |
dc.subject.keyword | 比例控制,模糊控制,B-spline curve,視覺系統,路徑追蹤,障礙物迴避, | zh_TW |
dc.subject.keyword | Path-planning,Path-tracking,B-spline curve,Fuzzy Control,Collision Cone,real-time, | en |
dc.relation.page | 67 | |
dc.identifier.doi | 10.6342/NTU202100682 | |
dc.rights.note | 未授權 | |
dc.date.accepted | 2021-02-14 | |
dc.contributor.author-college | 工學院 | zh_TW |
dc.contributor.author-dept | 應用力學研究所 | zh_TW |
顯示於系所單位: | 應用力學研究所 |
文件中的檔案:
檔案 | 大小 | 格式 | |
---|---|---|---|
U0001-0802202116012600.pdf 目前未授權公開取用 | 6.66 MB | Adobe PDF |
系統中的文件,除了特別指名其著作權條款之外,均受到著作權保護,並且保留所有的權利。