基礎設施建築物的劣化為著重的問題,最主要的原因在於全球暖 化、極端氣候及工程品質不穩定,導致每座橋梁都必須做例行性維護 與檢測;但因勞動力人口下降,導致目前檢測逐漸趨於自動化,而無 人飛行機無非是現今最佳選擇,然而飛行機會因馬達轉速差異,導致 機身在移動時產生傾斜,進而影響到照片品質。 針對照片傾斜問題,本論文希望開發一種新型飛行機,藉由向量 控制去優化原飛行機移動所產生的傾斜問題,以利於在進行建築物檢 測、拍攝時,更快速、效率收集高品質照片。 向量控制飛行機,主要包含機體穩定、機構設置與優化照片,依 據各馬達做X、Y 方向控制,並估算穩定方程式;在機構部分,則以 機臂做改良,以達到向量控制的機能;最後會比對在有無向量控制的 情況下,是否能收集到高品質照片。
Infrastructure deterioration is one of the biggest problem in engineering, thus, regular inspection of the bridge is a must. Traditional inspection method involving human inspector using aerial ladder, which is dangerous and most importantly, the inspection visual only. For these reason, many enterprise and university are using unmanned aerial vehicle to capture aerial image/video to perform building inspection. However, all vehicle suffer attitude change when performing translation x and y direction, which effect the qualities of photos. This thesis aims at tackle the problem to increase the photo qualities by removing attitude change. The method that will be applied is vector thrust propulsion system, which instead of attitude change in vehicle, the motor will change its attitude to allow the vehicle to maneuver to perform building inspection and collect image data. By constructing conventional multi-rotor vehicle mathematical model as starting point, this thesis will focus on the modeling of vector thrust and actual mechanism for the multi-rotor vehicle. Hardware realization will be based on the modification of “motor arm” in the existing frame, and lastly, the inspection experiment will be carry out to compare image qualities between conventional design and vector thrust design.