隨著科技蓬勃發展及無人載具應用於各領域的需求與日俱增,與無人飛行載具相關之研究方興未艾;而自動降落對於無人飛行載具而言,是一個重要且充滿挑戰的關鍵性階段。由於無人直昇機具有高機動性、垂直起降、定點停懸等優勢;嵌入式系統研發日新月異,以高效能、高可靠性、低功耗、低成本、體積小、重量輕為訴求;且近年來機器視覺應用多元化,可作為以非接觸、非破壞的方式取得環境影像資訊的感測器。有鑒於此,本研究之無人直昇機規劃使用嵌入式控制器作為開發平台,搭載單台攝影機為機器視覺感測器擷取即時的影像資訊,以影像處理演算法的程式開發為手段,透過降落圖示與影像間座標點的對應關係,估測載具與降落圖示間的相對距離與姿態。經實驗結果證明,本研究之攝影機標定方法以單張降落圖示影像估測載具與降落圖示間的相對距離與姿態,由影像座標間的關係得知估測結果與實際趨勢相符,而本研究所估測之載具飛行高度較原先以圖示面積進行高度查表的方式更符合實際高度;在定位方面亦可提供比誤差範圍在公尺等級的定位系統更準確之平移資訊;但由於相對姿態的估算結果較不穩定,在某些角度下會超出容許範圍,為考量實際無人載具飛行之穩健性,目前仍無法取代現有的姿態感測器,並對於實驗結果不如預期之原因進行探討,提出未來可以改良的方向,進而達到導引無人直昇機自動降落之目的。
Recently, the demand of unmanned aerial vehicle(UAV) becomes more popular in many fields, and automatic landing is an important and challenging issue for UAV. Because of the advantages of the unmanned aerial helicopter(UAH), such as high mobility, vertical takeoff and landing, and hovering mode, this study proposed to develop the automatic landing guidance system for unmanned aerial helicopter. Moreover, the machine vision has been used in many systems to retrieve the image information of environment with noncontact and nondestructive methods. This is helpful to get the information of the attitudes and location of the unmanned aerial helicopter. On the other hand, embedded system is ideal for the development of UAV flight system. Because there are several features, for instance, small size, light weight, low power consumption, highly processing speed and stability. The purpose of this study is to develop a machine vision system which can provide the estimated attitudes and relative distance to the designated landing position of unmanned aerial helicopter for automatic landing. In this study, a digital camera is used to identify the landing mark with an embedded system hardware architecture. Through the techniques of image processing, this study develops the required program to calculate the relationship of the image and landing mark.