跳到主要內容

臺灣博碩士論文加值系統

(44.200.140.218) 您好!臺灣時間:2024/07/26 00:13
字體大小: 字級放大   字級縮小   預設字形  
回查詢結果 :::

詳目顯示

我願授權國圖
: 
twitterline
研究生:周子翔
研究生(外文):Chou,Tzu-Hsiang
論文名稱:視覺導引多軸無人機海上甲板降落系統
論文名稱(外文):Visually Guided Landing System on Ship Deck for Multicopter
指導教授:李東霖李東霖引用關係
指導教授(外文):Li, Dong-Lin
口試委員:莊仁輝蔡宇軒李東霖
口試委員(外文):Chuang, Jen-HuiTsai, Yu-ShiuanLi, Dong-Lin
口試日期:2022-01-21
學位類別:碩士
校院名稱:國立臺灣海洋大學
系所名稱:電機工程學系
學門:工程學門
學類:電資工程學類
論文種類:學術論文
論文出版年:2022
畢業學年度:110
語文別:中文
論文頁數:77
中文關鍵詞:多軸旋翼機海上降落電腦視覺ArUco標籤類神經網路
外文關鍵詞:MulticopterLanding on Ship DeckComputer VisionArUco MarkerNeural Networks
相關次數:
  • 被引用被引用:0
  • 點閱點閱:148
  • 評分評分:
  • 下載下載:0
  • 收藏至我的研究室書目清單書目收藏:1
傳統海上搜尋高價值魚群主要是透過直升機,而隨著科技發展,海上無人機的發展在近年來日益重要,相較於傳統直升機方法,無人機的飛行成本大幅降低,因此無人機取代傳統直升機方法已經是種趨勢。但由於海浪的拍打再加上海上的陣風,使得船隻搖晃讓無人機難以降落。
因此本論文提出一種更穩定的多軸旋翼機降落於海上船隻方法,包含在更準確的影像標籤抓取、更穩定的抗風控制與更精準的甲板晃動預測。在影像標籤抓取部份本論文提出一種在單張低解析度影像下提昇Aruco標籤的輪廓角點抓取精準度方法。而穩定的抗風控制部份是根據在世界坐標系與ArUco標籤的誤差當作輸入,透過PID控制器計算出的速度分量控制無人機。在甲板晃動預測部份是經由在線倒傳遞神經網路(Backpropagation Neural Network)預測船隻Roll的傾角,以預測出最適合降落的時間。本論文不但經由陸上模擬實驗與其它方法比較證明其較佳的效果,最重要的是在真實海上環境進行降落實際驗證。
Traditionally, the search for high-value fish at sea is mainly through helicopters. With the development of technology, the development of maritime drones has become increasingly important in recent years. Compared with traditional helicopter methods, the flight cost of drones is greatly reduced, so drones replace traditional helicopter methods. Already a trend. But the pounding waves and gusts of wind over the sea made it difficult for the drone to land as the boat swayed.
Therefore, this paper proposes a more stable method for multi-axis rotorcraft to land on ships at sea, including more accurate image tag capture, more stable wind resistance control and more accurate deck shaking prediction. In the image label capture part, this paper proposes a method to improve the accuracy of ArUco Marker's contour and corner capture in a single low-resolution image. The stable wind resistance control part is based on the error between the world coordinate system and the ArUco Marker as input, and controls the drone through the velocity component calculated by the PID controller. In the sway prediction part of the deck, the inclination angle of the ship's Roll is predicted through the online Backpropagation Neural Network to predict the most suitable time for landing. This paper not only proves its better effect by comparing the land simulation experiment with other methods, but also the most important is the actual verification of landing in the real sea environment.
摘要I
AbstractII
致謝III
目次IV
圖目次VI
表目次IX
第一章 緒論1
1.1研究動機1
1.2 問題與挑戰2
1.3相關文獻介紹4
1.3.1標籤相關研究4
1.3.2國內外相關研究6
1.3.3小結11
1.4研究目標12
第二章 相關研究13
2.1相機校正13
2.2影像定位16
2.2.1 ArUco Marker16
2.2.2 亞像素細化20
2.3角度計算21
2.3.1 Perspective-n-Point21
2.3.2倒傳遞神經網路24
第三章 研究方法26
3.1系統架構26
3.2定位與抗風28
3.2.1定位28
3.2.2 角點細化30
3.2.3 抗風33
3.3角度預測36
第四章 實驗結果38
4.1 實驗環境介紹38
4.1.1 硬體介紹38
4.1.2實驗環境設定40
4.2陸上實驗結果47
4.2.1精準降落實證47
4.2.2角度預測實驗結果49
4.2.3抗風效果實證51
4.3海上模擬驗證54
4.3.1船隻平移運動模擬54
4.3.2船隻升沉運動模擬56
4.4海上降落驗證57
第五章 結論與未來展望61
5.1結論61
5.2 未來展望62
參考文獻63
[1] 行政院,“漁業統計年報”,2016. ,[Online]. Available: https://www.fa.gov.tw/cht/PublicationsFishYear/ .
[2] 高雄市政府海洋局, “船筏種類” ,[Online]. Available: https://kcmb.kcg.gov.tw/cp.aspx?n=7988CDDC702EB309.&Create=1.
[3] esentraAdmin, “電力無人機巡檢中的RTK 技術,” 先創國際, Apr. 09, 2019. https://www.esentra.com.tw/2019/04/電力無人機巡檢中的rtk-技術/
[4] L. Chamberlain, S. Singh, and S. Scherer, “Self-Aware Helicopters: Full-Scale Automated Landing and Obstacle Avoidance in Unmapped Environments,” 2011, [Online]. Available: https://www.semanticscholar.org/paper/Self-Aware-Helicopters%3A-Full-Scale-Automated-and-in-Chamberlain-Singh/a0f75f08588adf01a616a5568a46d32a4f03dcf5.
[5] E. Olson, “AprilTag: A robust and flexible visual fiducial system,” in 2011 IEEE International Conference on Robotics and Automation, May 2011, pp. 3400–3407. doi: 10.1109/ICRA.2011.5979561.
[6] S. Garrido-Jurado, R. Muñoz-Salinas, F. J. Madrid-Cuevas, and M. J. Marín-Jiménez, “Automatic generation and detection of highly reliable fiducial markers under occlusion,” Pattern Recognition, vol. 47, no. 6, pp. 2280–2292, Jun. 2014, doi: 10.1016/j.patcog.2014.01.005.
[7] B. Benligiray, C. Topal, and C. Akinlar, “STag: A Stable Fiducial Marker System,” arXiv:1707.06292 [cs], Jul. 2019, [Online]. Available: http://arxiv.org/abs/1707.06292.
[8] 陳建綸, “基於視覺導航修正全球定位系統之旋翼機自動定點降落,” 國立成功大學, 台南市, 2019. [Online]. Available: https://ndltd.ncl.edu.tw/cgi-bin/gs32/gsweb.cgi/ccd=nlbORn/record?r1=1&h1=0.
[9] 邱育良, “基於視覺定位之多軸飛行器降落,” 國立聯合大學, 苗栗縣, 2020. [Online]. Available: https://ndltd.ncl.edu.tw/cgi-bin/gs32/gsweb.cgi/ccd=WcxzRo/record?r1=1&h1=0
[10] 劉効哲, “無人機於建築物周圍指定區域之視覺導航降落方法,” 國立政治大學, 台北市, 2020. [Online]. Available: https://ndltd.ncl.edu.tw/cgi-bin/gs32/gsweb.cgi/ccd=UX4kRs/record?r1=1&h1=0
[11] J. Wynn, “Visual Servoing for Precision Shipboard Landing of an Autonomous Multirotor Aircraft System,” Brigham Young University, State of Utah, 2018. [Online]. Available: https://scholarsarchive.byu.edu/etd/7111.
[12] J. Wubben et al., “Accurate Landing of Unmanned Aerial Vehicles Using Ground Pattern Recognition,” Electronics, vol. 8, no. 12, Art. no. 12, Dec. 2019, doi: 10.3390/electronics8121532.
[13] S. Abujoub, J. McPhee, and R. A. Irani, “Methodologies for landing autonomous aerial vehicles on maritime vessels,” Aerospace Science and Technology, vol. 106, p. 106169, Nov. 2020, doi: 10.1016/j.ast.2020.106169.
[14] B. Li, B. Wang, X. Tan, J. Wu, and L. Wei, “Corner location and recognition of single ArUco marker under occlusion based on YOLO algorithm,” JEI, vol. 30, no. 3, p. 033012, May 2021, doi: 10.1117/1.JEI.30.3.033012
[15] 拾人牙慧,” Pinhole Camera: 相機校正 (Camera Calibration)”, 2016, [Online]. Available: http://silverwind1982.pixnet.net/blog/post/153218861.
[16] 維基百科, " Perspective-n-Point," Dec. 22, 2021. [Online]. Available: https://en.wikipedia.org/w/index.php?title=Perspective-n-Point&oldid=1061558184.
[17] V. Lepetit, F. Moreno-Noguer, and P. Fua, “EPnP: An Accurate O(n) Solution to the PnP Problem,” Int J Comput Vis, vol. 81, no. 2, p. 155, Jul. 2008, doi: 10.1007/s11263-008-0152-6.
[18] "Applications of Artificial Vision" (A.V.A) of the University of Cordoba, “ArUco: a minimal library for Augmented Reality applications based on 2022OpenCV”, [Online]. Available: http://www.uco.es/investiga/grupos/ava/node/26.
[19] OpenCV, ” Detection of ArUco Markers”, [Online]. Available: https://docs.opencv.org/4.x/d5/dae/tutorial_aruco_detection.html.
[20] 學習堅持,堅持學習, “Back Propagation Neural Network 倒傳遞類神經網路,”[Online]. Available: https://dotblogs.com.tw/dragon229/2013/01/23/88750.
[21] “PID控制器,” 維基百科,自由的百科全書. Feb. 08, 2021. [Online]. Available:https://zh.wikipedia.org/w/index.php?title=PID%E6%8E%A7%E5%88%B6%E5%99%A8&oldid=64184356
[22] 王星宇, “外海可沉式箱網養殖區域劃定初探”, 2018. [Online]. Available: http://www.tsoe.org.tw/downloads/thesis/2018G502.pdf.
[23] “蒲福氏風級,” 維基百科,自由的百科全書. Dec. 14, 2021. [Online]. Available:https://zh.wikipedia.org/w/index.php?title=%E8%92%B2%E7%A6%8F%E6%B0%8F%E9%A2%A8%E7%B4%9A&oldid=69095638.
[24] PX4 User Guide, “Pixhawk 4 Mini | PX4 User Guide.”, [Online]. Available: https://docs.px4.io/master/en/flight_controller/pixhawk4_mini.html.
[25] NVIDA, ”Jetson Nano”, [Online]. Available: https://www.nvidia.com/zh-tw/autonomous-machines/embedded-systems/jetson-nano/.
[26] R. P. Ltd, “Buy a Raspberry Pi Camera Module 2,” Raspberry Pi, [Online]. Available: https://www.raspberrypi.com/products/camera-module-v2/
[27] 翔大模型, “新款輕量145克免調2軸雲臺”, [Online]. Available: https://www.ruten.com.tw/item/show?21637634834311.
[28] “Vessel Motion Calculator | displacement velocity acceleration.”, [Online]. Available: https://www.calqlata.com/productpages/00059-help.html
[29] F. TAN, “ship motion.” IEEE, Apr. 05, 2020. [Online]. Available: https://ieee-dataport.org/documents/ship-motion.
[30] B. Zhou, X. Qi, J. Zhang, and H. Zhang, “Effect of 6-DOF Oscillation of Ship Target on SAR Imaging,” Remote Sensing, vol. 13, no. 9, Art. no. 9, Jan. 2021, doi: 10.3390/rs13091821.
[31] 交通部中央氣象局, “臺灣近海海況” , Oct. 25, 2021., [Online]. Available: https://www.cwb.gov.tw/V8/C/M/NSea.html.
電子全文 電子全文(網際網路公開日期:20270216)
連結至畢業學校之論文網頁點我開啟連結
註: 此連結為研究生畢業學校所提供,不一定有電子全文可供下載,若連結有誤,請點選上方之〝勘誤回報〞功能,我們會盡快修正,謝謝!
QRCODE
 
 
 
 
 
                                                                                                                                                                                                                                                                                                                                                                                                               
第一頁 上一頁 下一頁 最後一頁 top
無相關期刊