利用影像辨識輔助無人飛機自動降落之策略研究

The Research of Strategy Use Image Recognition to Assist UAV in Auto Landing

林偉傑(Wei-Chieh Lin)

指導教授 ： 吳昌暉

元智大學/工程學院/機械工程學系/碩士(2013年)

https://doi.org/10.6838/YZU.2013.00157

摘要

無人飛行載具（Unmanned Aerial Vehicle，UAV），近年來在軍事以及民生上的運用越來越廣泛，但至今仍無法普及的一大原因在於降落的困難度。而各大機場自動降落的儀器著陸系統，設備價格不菲且場地受限，因此欲開發一套以影像辨識為基礎的自動降落輔助系統。
本論文主要為改良本實驗室所建立適用於無人飛機之影像辨識降落輔助系統，此系統可針對所擷取之即時影像，計算得機體與跑道的相對資訊，將橫向偏移量轉為修正命令，結合降落導引控制律，可彌補GPS的精度誤差，從12公尺修正至3公尺內，處理速度約15fps，以HIL完成自動降落模擬實驗。

關鍵字

無人飛機 ； 自動降落 ； 影像辨識 ； 控制律

並列摘要

This paper improves the Image Recognizing Landing Assistance System mainly. This is a system through image processing method to detect the landing runway from real-time video. Found the centerline of runway and calculated the relative information by image recognition.
In this paper, the author chooses the suitable recognition target, advances algorithm more robust and more effective, achieves the data transmission. Use HIL processing the auto landing experiment that prove this system fixing the GPS error from 12 meters within 3 meters; it can make the aircraft approach the middle line of runway.

並列關鍵字

UAV ； Image Recognition ； Auto Landing

參考文獻

10. 無人飛機影像辨識輔助降落系統，元智大學，張展維，民國101

1. A Vision System for Landing an Unmanned Aerial Vehicle, University of California, Courtney S.Sharp Orriid Shakernia S.Shankar Sastry, 2001

2. Vision Based Terrain Recovery for Landing Unmanned Aerial Vehicles, University of California, Berkeley, Marci Meingast, Christopher Geyer, and Shankar Sastry, 2004

3. Vision Assisted Autonomous Landing of an Unmanned Aerial Vehicle. Vilas K.Chitrakaran, Darren M.Dawson, Jian Chen, and Matthew Feemster Member, IEEE, 2005

4. Landing a UAV on a Runway Using Image Registration, University of California, Andrew Miller, Mubarak Shah, Don Harper, 2008