论文部分内容阅读
光纤陀螺的随机漂移限制了惯性导航系统的精度,如何减小它是一项非常艰巨的任务。结合经验模态分解(EMD)和信号与模态之间的概率密度函数,提出了一种新型的依赖Hurst指数的信号滤波方法。当H<0.5时,利用l_2范数选择出相关模态,累加并形成的部分重构方法来对光纤陀螺的信号进行滤波;当H≥0.5时,间隔阈值的经验模态(EMD-IT)被引入对相关模态进行滤波,之后按照部分重构的方法对光纤陀螺的信号进行滤波;称为混合的EMD-pdf和EMD-IT。与其它的滤波方法进行对比,如基于相关函数的EMD部分重构(EMD-cor),基于概率密度函数的EMD部分重构(EMD-pdf),仿真信号和实际数据结果表明,该混合模型的优越性,有效减小了光纤陀螺的随机误差。
The random drift of FOG limits the accuracy of inertial navigation system. How to reduce it is a very difficult task. Combining empirical mode decomposition (EMD) and the probability density function between signal and mode, a new signal filtering method based on Hurst exponent is proposed. When H <0.5, the mode of l_2 norm is used to select the relevant modalities, and the partial reconstructed methods are accumulated and formed to filter the signal of the fiber optic gyroscope. When H≥0.5, the empirical mode of the interval threshold (EMD-IT) Was introduced to filter the relevant modalities and then filtered the signal from the fiber optic gyroscope in a partially reconstructed manner; this was called a hybrid EMD-pdf and EMD-IT. Compared with other filtering methods, such as EMD partial reconstruction (EMD-cor) based on correlation function, EMD partial reconstruction (EMD-pdf) based on probability density function, the simulation signal and actual data show that the hybrid model The superiority effectively reduces the random error of the FOG.