擴增型卡曼濾波器應用於微型振動陀螺儀之即時控制系統

Abstract

本篇論文中，提出了使用在振動陀螺儀上，估測角速度的一個新方法。以往大部分的作法，都是假設系統量測雜訊可忽略，並且在其他參數都是已知的前提之下進行估測。不同於以往，我們考慮量測訊號存在雜訊，及所有參數未知的情況下，利用「擴增型卡曼濾波器(extended Kalman filter)」作為觀察器的演算法正確地估測出角速度，並鑑別系統參數。 由MATLAB軟體的模擬結果顯示本方法在七個參數未知的情況下，及量測雜訊 下(訊噪比約為 )，可鑑別出所有參數，同時角速度的估測精確度仍可達 。此外，透過觀察器的狀態回授控制，可將系統結構存在的阻尼及「耦合力(coupling force)」消除，經由求解控制過後的系統動態方程式，我們可以直接計算出角度，避免因積分角速度而產生的誤差累積。

In this thesis, we present a novel approach to estimate angular rates of MEMS vibratory gyroscope systems. Different from most of the existing methods, wherein all system parameters were assumed to be known and system measurements were noise-free, this method uses extend Kalman filter (EKF) techniques to estimate all the system parameters, including the angular rate. And, it nulls out the noise from system measurements and thus increases the sensing accuracy for gyroscope systems. Simulation results indicate that the proposed method can estimate all the system parameter correctly. The estimation accuracy of angular rate is when the standard deviation of the measurement noise is (signal-to-noise ratio is about twenty). In addition, the proposed method can be used to compensate the damping terms and “coupling forces” in gyroscope systems. Subsequently, the rotation angles can be calculated directly. As compared to the conventional methods wherein the angles were obtained by integrating angular rates over time, the proposed method does not suffer from the error accumulations and thus more accurate.