- 學位論文
四旋翼飛行器自動泊接系統開發
Development of an Automatic Docking System for a Quadcopter
摘要
多年來，四旋翼飛行器(簡稱四旋翼)在無人飛行載具領域上是一個熱門的研究對象。而這幾年，四旋翼更已高度商品化，從遙控玩具到專業空拍活動，都有四旋翼活躍的身影，其應用和周邊商品也越來越多元。 在無人飛行載具的應用及研究，不管四旋翼無人飛行器是由人為控制或者是飛行器自主飛行，四旋翼的續航力是影響四旋翼活動範圍及航行時間的關鍵因素，也是值得探討的議題。市面上常見的四旋翼無人飛行器續航力約為15~30分鐘，接近操作時間上限就必須使四旋翼返航充電；充電場所因為環境以及充電設備的關係，室內是一個較好的選擇。本研究基於上述四旋翼飛行器續航力不足以及四旋翼須在室內進行充電兩點，設計一個四旋翼自動泊接平台，提供四旋翼在室內能停靠的平台，未來能在平台上實現自動充電的功能。 由於室內無法接收到GPS訊號，所以本研究利用光學定位方式來完成四旋翼於室內的定位。實作中把任天堂Wiimote的紅外線感測鏡頭架設於懸空平台作為光學定位的感測模組，追蹤四旋翼載具上的紅外線光源，依據裝設在四旋翼機體的紅外線光源於Wii紅外線鏡頭成像面所組成的三角形，透過透視投影原理以及簡單的幾何計算，估算四旋翼和Wii紅外線鏡頭在空間中的相對位置；在四旋翼動作控制方面，本研究以Arduino做為核心，除了處理Wii紅外線鏡頭的感測圖像，估算四旋翼的位置，同時Arduino也根據估算結果與預設目標的差異值產生四旋翼飛行控制命令，再以無線方式控制四旋翼的飛行動作。 本研究已經成功使四旋翼在進入Wiimote紅外線感測鏡頭視野範圍之後，能自主修正速度與位置，並向上接近並泊停至預設平台，完成四旋翼自動停靠的功能。
並列摘要
In this decade, quadcopters have become a popular research topic in the field of Unmanned Aerial Vehicles (UAV). In the application and research of UAVs, regardless of quadcopters are controlled manually or fly automatically, the endurance of quadcopters is the key factors which affecting quadcopters' operational time, and it is also the point which is worthy to be discussed. Although quadcopters are studied by lots of people, there are two main issues of the endurance of quadcopters. First, the endurance of quadcopters is insufficient. The endurance of most commercial quadcopters is 15 to 30 minutes. It means that quadcopters can be used only in a short time, and they have to perform recharging frequently. Second, quadcopters must charge the battery indoor under the constraints of the environment and equipment. Based on these two points, we propose a prototype of an automatic docking system for indoor operation. Moreover, we hope quadcopters to perform automatic charging on the platform in the future. In this research, a Wiimote IR camera is installed on a platform to track the positioning signals from IR LEDs on a quadcopter. The relative position between the quadcopter and the Wiimote IR camera can be estimated by an embedded system, Arduino, which based on projected values of pixel coordinate from three IR signals. After that, the embedded system generate and send the control signal to control the quadcopter wirelessly. In this research, we have successfully realized a quadcopter to perform automatic approaching and docking to the docking platform in the field of view of the IR camera.