基于卡尔曼滤波的陀螺仪降噪处理

针对MEMS陀螺仪输出信号随机漂移误差造成测量精度低的问题,提出了一种基于BP神经网络的卡尔曼滤波降噪模型.基于BP神经网络的基本原理,首先利用BP神经网络对系统进行学习,获得系统状态方程,然后建立了基于BP神经网络的滤波模型,最后应用于卡尔曼滤波对MEMS陀螺仪信号进行降噪.半实物模拟仿真实验表明:基于BP神经网络的卡尔曼滤波后的数据的速率随机游走等系数比原始数据下降6.89倍,验证了本方法的降噪性能优于基本卡尔曼模型,在MEMS陀螺仪的数据处理方面具有一定的应用价值.

The noise reduction of gyroscope based on Kalman filter

Based on the requirement of MEMS gyroscope to improve output accuracy and reduce random drift error,a Kalman noise reduction model based on BP neural network is built. The basic principle is introduced. Firstly,the BP neural network is used for learning to obtain accurate state equation of the system. Then,the filtering model based on BP neural network is established. Finally,the model is applied to Kalman filter to denoise the MEMS gyro signal. The semi-physical simulation experiments indicate the random walk of rate coefficient of Kalman filter data based on BP neural network model can be decreased by 6.89 times compared to the original data. This method has better noise reduction performance than the basic Kalman model and has certain application value in data processing of MEMS gyroscope.