[[alternative]]Integrated Flight Control System Simulation with Saturating Controller Synthesis and Terrain Contour Matching

[[abstract]]本計劃主要目的在於將會飽和制動器之設計, 地形比對與飛行模擬系統整合, 俾能使飛行器依預先儲存之地形資料庫修正慣性導航系統(INS)在長程飛行下所造成的誤差。計劃內容主要包含下列三項工作要點:(1)在地形比對:本研究使用二維數位影像地圖來做實測影像與參考影像之間的比對, 並且正確的找出實測影像在參考影像上的位置。主要任務是考慮此辨認法之抗旋轉(Rotation invariant)特性, 作法分為兩個部份:第一, 先在頻域上做粗尋, 在影像上取一圓形的面積並分別對此圓面積的徑向R與角度.theta.兩個方向來抽取影像像素值(Pixels)存成一個M*N的二維陣列, 並對此二維陣列做快速傅立葉轉換(FFT)得到影像的特徵, 用此特徵來比對找出目標粗略位置。第二, 在粗尋所找出的影像範圍中, 回到時域上做細尋比對。(2)會飽和制動器之設計:作者提出一Lyapunov function作為控制器設計的成本函數。利用此函數首次吾人可證明出一含雙積分(Double integrators)的系統可被一會飽和的線性狀態回饋來全域(Globally)穩定此類函數尚未見諸於其他論文中。(3)飛行系統模擬:吾人在 MATLAB/SIMULINK環境下, 將系統區分為快速反應次系統和慢速反應次系統。而後利用多重積分率達到快慢系統間積分的匹配, 以節省計算時間。[[abstract]]The main purpose of this project is to integrate the saturating controller synthesis, terrain contour matching, and flight simulation. (1) Terrain matching: we provide an anti-rotation terrain matching method. We used a two- dimensional digital terrain image to perform terrain matching and locate the correct position of the sensed image. This Method can be divided into two steps. First, a two-dimensional Fast Fourier Transform (2DFFT) is used to perform image matching in frequency domain, and find the proper area that contains the sensed image. Second, return to time domain to do an image matching in the previous chosen area. (2) The synthesis of a saturating controller: the author present a Lyapunov function for a system (may contain double integrator) with saturating controller . By using this Lyapunov function, we for the first time show that a system containing a double integrator can be globally stabilized by a linear saturating controller. (3) Flight system simulation: Under the environment of MATLAB/SIMULINK the system is divided into fast and slow subsystem. Then a multiple rate integration method is applied to reduce computing time.[[conferencetype]]國內[[conferencetkucampus]]淡水校園[[conferencedate]]19980516~19980516[[iscallforpapers]]Y[[conferencelocation]]臺北縣, 臺