基于优化KF的MEMS陀螺阵列信号融合方法

针对微机电系统(micro electro mechanical system,MEMS)陀螺仪准确度低、噪声大的问题,采用陀螺阵列技术降噪以提高陀螺的使用精度。采用Allan方差法分析陀螺信号误差噪声项,依据分析结果对测量模型进行了简化,利用噪声相关性设计了一种卡尔曼滤波器(Kalman filter,KF)对陀螺阵列进行数据融合,并对最优估计过程进行了改进,降低了数据处理的复杂度和计算量,同时从理论上分析了各参数对阵列性能的影响。为提高滤波器的动态性能,将自回归(autoregressive,AR)模型应用于陀螺真实角速率的建模。采用6个陀螺构成的阵列进行了验证实验。实验结果表明:与单个陀螺相比,陀螺阵列的噪声在静态条件下降低了144.2倍,在恒速率和正弦速率条件下噪声分别降低了18.18倍和5.36倍,证明了此建模方法和融合方法的有效性。

Fusion method of MEMS gyro array signals based on optimal KF

To solve the problem of low accuracy and great noise, the gyro array technique is used to reduce noise and improve the performance of micro electro mechanical system (MEMS) gyroscope. The measurement model is simplified on the basis of the component of noise in MEMS gyroscope, which is analyzed by Allan variance. Furthermore, a novel Kalman filter (KF) for combining outputs of a gyroscope array using cross correlation between noises from different gyroscopes is designed. The optimal estimation process is improved to decrease the complexity and calculation quantity, then the accuracy of gyro array and affecting factors are analyzed by using a steady state covariance. Considering the dynamic performance of the signal, the true rate model is established by the autoregressive (AR) model. The experimental results indicate that the precision of the gyro array composed by six gyroscope increases 144.2 times in static condition. In dynamic condition, the precision increases 18.18 times when the input rate is constant and 5.36 times when input rate is sine function. It proves the validity of the modeling and fusion methods.