English  |  正體中文  |  简体中文  |  Items with full text/Total items : 26994/38795
Visitors : 2392545      Online Users : 59
RC Version 4.0 © Powered By DSPACE, MIT. Enhanced by NTU Library IR team.
Scope Adv. Search
LoginUploadHelpAboutAdminister

Please use this identifier to cite or link to this item: http://ntour.ntou.edu.tw:8080/ir/handle/987654321/30378

Title: 以操船模擬機為測試平台之影像導引船舶避碰系統探討
Use of a shiphandling simulator as the test platform in the study of an image processing-based ship collision avoidance system
Authors: Zhao-Heng Huang
黃肇珩
Contributors: NTOU:Department of Communications Navigation and Control Engineering
國立臺灣海洋大學:通訊與導航工程學系
Keywords: 影像處理;疊標導航;船舶避碰;模糊控制;操船模擬機
Image Processing;Leading Line Guidance;Ship Collision Avoidance;Fuzzy Control;Shiphandling Simulator
Date: 2011
Issue Date: 2011-11-25T07:43:39Z
Abstract: 本文利用影像處理技術,探討船舶沿特定導航疊標線入港過程中,遭遇固定或移動障礙物時利用影像處理技術計算本船與障礙物之相對方位及距離資訊,提供模糊自航器完成避碰操演。實驗之進行乃透過架設於模擬機駕駛台外側框架上方之CCD攝影機,擷取置於障礙物(他船)之目標物,並經由影像處理計算目標物質心位置,以做為估算本船與他船相對方位及距離之依據。當本船完成避碰操演後,藉由岸上導航疊標之幾何中心及影像座標關係,可提供做為本船和導標間的航向角修正量,該資訊提供給模糊航向控制器即可將船舶再次導引回原預訂航道線上。實驗藉由CCD攝影機下方之伺服馬達構成旋轉機制,有效降低了因視角侷限所造成的出鏡問題。本實驗使用LABVIEW / Vision Build AI 工作環境,在影像處理技術上,採取顏色辨別及中值濾波、型態學閉合達到目標物選定,實驗數據顯示,於無海況條件下,利用影像處理技術之平均距離估測誤差約在10%之內。而本研究之特色為利用海洋大學操船模擬機環境設定及模擬便利性,因此得以有效率的進行各項干擾(風、流、海浪)之實驗測試。模擬實驗結果顯示,於風、流外力干擾下,藉由影像辨識針對固定障礙物及移動障礙物之船舶航行自動避碰確實具有其可行性。
This work is on the application of image processing technique and fuzzy logic control to the design and implementation of a shiphandling simulator-based ship collision avoidance system. Specifically, during the approaching maneuver, the own ship is required to avoid an obstacle ship that might cross the ownship’s intended track which is typically defined by a leading line, and the ownship should resume its original approaching mode, immediately after completion of the collision avoidance maneuver. A CCD camera mounted on top of the frame outside of the simulator is used in capturing images of a set of leading marks placed on the shore during the approaching maneuver and a set of targets placed on an obstacle ship during the avoidance maneuver. The centers of the target images are computed by an image processing unit and the relative orientation and the distance between the ownship and the targets are fed into the fuzzy autopilot to initiate necessary maneuvers. The Labview Vision Builder AI environment is adopted and the color recognition, median filtering and closing operations are used in acquiring the target images. Experimental data indicates that under zero sea state condition distance estimation error by using the image processing technique is about 10%. Placement of a rotating stand under the CCD successfully deals with the limited view angle problem associated with the traditional CCD camera. A series of successful collision avoidance maneuvers conducted on the NTOU SHS (National Taiwan Ocean University Shiphandling Simulator) under disturbance effects indicates the feasibility of the proposed ship collision system basing on image processing technique and fuzzy logic control algorithm.
URI: http://ethesys.lib.ntou.edu.tw/cdrfb3/record/#G0M98670034
http://ntour.ntou.edu.tw/handle/987654321/30378
Appears in Collections:[通訊與導航工程學系] 博碩士論文

Files in This Item:

File Description SizeFormat
index.html0KbHTML256View/Open


All items in NTOUR are protected by copyright, with all rights reserved.

 


著作權政策宣告: 本網站之內容為國立臺灣海洋大學所收錄之機構典藏,無償提供學術研究與公眾教育等公益性使用,請合理使用本網站之內容,以尊重著作權人之權益。
網站維護: 海大圖資處 圖書系統組
DSpace Software Copyright © 2002-2004  MIT &  Hewlett-Packard  /   Enhanced by   NTU Library IR team Copyright ©   - Feedback