【作者】 熊林香；

【导师】 潘英俊；

【作者基本信息】 重庆大学 ， 光学工程， 2010， 硕士

【摘要】 捷联惯性导航系统因其具有完全自主性和不受干扰的优势在军民领域尤其是军用领域得到了广泛的应用。捷联惯性导航系统精度的高低是衡量一个国家军事实力的重要标志之一。影响捷联惯性导航系统精度的因素很多,比如惯性敏感元件(陀螺和加速度计)的测量精度、信号处理的精度和速度以及姿态解算的算法等等。因此,在硬件实现和算法选择相同的条件下,惯性敏感元件(陀螺和加速度计)的测量精度就变得尤为重要。零漂是衡量陀螺仪性能的一个重要指标。目前,针对零漂的处理方式大多是对其进行建模,然后根据模型进行补偿,但是陀螺的漂移往往是弱非线性、非平稳、慢时变的,必须在线实时拟合模型、辨识参数才能达到补偿效果,这在实时系统中难于实现。针对目前捷联惯性导航系统高精度的需求,本文基于目前广泛使用的惯性敏感元件光纤陀螺,在其零漂信号的分析与处理方面开展了如下的研究工作:①对实验室的某型号光纤陀螺进行了测试,得到其零漂信号。采用Allan方差对光纤陀螺实测零漂信号进行分析,得到各噪声系数,再根据各噪声的形成机理找到光纤陀螺的噪声源,为提高光纤陀螺的精度提供参考价值。②利用小波变换阈值滤波对光纤陀螺的零漂信号进行了直接滤波处理。研究了小波变换阈值滤波在光纤陀螺零漂信号处理中小波基、阈值、阈值作用方式和分解层数的选取原则,重点研究了小波基和阈值的选取。通过仿真试验得到了最佳的滤波方案。③利用Allan方差得到光纤陀螺零漂信号滤波处理前和不同阈值条件下采用小波变换阈值滤波后各噪声系数的大小,定量地验证了滤波的效果。为光纤陀螺和其他类陀螺的零漂信号的处理提供了参考价值,从而提高其工程实用能力。更多还原

【Abstract】 With full autonomy and uninterrupted advantages, strapdown inertial navigation system is widely used in the area of military and civilian, especially in military area. The accuracy of strapdown inertial navigation system is one of the important indicators of a country’s military strength. There are many factors which affect the accuracy of strapdown inertial navigation system, such as the measurement accuracy of inertial sensor (gyroscope and accelerometer), signal processing speed and accuracy ,posture solver algorithms. Therefore, under the same algorithm selection and hardware conditions, it is the measurement accuracy of gyroscope and accelerometer that becomes a very important factor. Drift is an important performance parameter of gyroscope. However, now most method of processing drift is making the model, and then compensating it according to the model, which is difficult to achieve in real-time system.In this thesis, some work on the analysis and processing of the fiber optical gyroscope (FOG)’s drift signal have been done, which is based on the requirement of high-accuracy strapdown inertial navigation system. The following are the main content done in this thesis.①The static test of a FOG is made, and the drift is gotten. Allan variance is studied, and the analysis of drift is made with Allan variance. The different noise coefficients of drift can also be gotten by Allan variance. The various noise sources are be gotten according to the different noise formation mechanism, which can afford a reference value for improve the accuracy of FOG.②The drift signal of FOG is processed with wavelet transform threshold-value filtering. The selection principle of wavelet function, threshold-value, the action of threshold-value and decomposition level are studied, focus on the wavelet function and the threshold-value. The optimal filtering scheme is gotten through simulation.③The different noise coefficients of FOG drift before and after different threshold-values filtering are gotten by Allan variance, which quantitatively verify the filtering result. Therefore, it provides a reference value for FOG and other gyroscope’s drift signal processing, and enhances their practical engineering ability to work.更多还原