MEMS陀螺仪漂移和噪声的分析和补偿

对陀螺仪数据分析的传统方法是使用kalman滤波器做尾数据处理来降低随机误差,由于陀螺仪传感器随着外界环境的变化的影响会有非线性误差,传统的kalman滤波算法处理的是线性误差,因此引进了适用于非线性系统的EKF滤波.为了快速滤除系统在实际环境中产生的噪声,对传统的中值滤波算法进行了改进,降低其计算复杂度,提出差分-均值中值滤波法.本文首先使用阿伦(ALLAN)方差分析了陀螺仪的误差特性,对于这些误差源分别提出了偏移校正的方法,之后建立自动回归-滑动平均模型(ARMA模型)对陀螺仪数据进行误差建模分析,最后使用EKF算法降低随机误差.实验结果表明该方法比传统的方法滤波效果好、计算复杂度低、实时性好.

Analysis and compensation of drift and noise in MEMS gyroscope

The traditional method of gyroscope data analysis is the tail data processing to reduce the random error u-sing the Kalman filter,the gyro sensor with the impact of changes in the external environment will have a nonlinear error,the introduction of EKF filtering for nonlinear systems.In order to rapidly filter the noise generated by the sys-tem in the actual environment,the traditional median filtering algorithm is improved to reduce its computational complexity,and a differential mean median filtering method is proposed. This paper first use Allen(ALLAN) analysis of variance of the error characteristics of gyroscope,error sources for these methods are proposed to offset correction,after the establishment of auto regressive moving average model(ARMA model)error modeling analysis of the gyroscope data,and finally use the EKF algorithm to reduce the random error.Experimental results show that the proposed method has better filtering effect,lower computational complexity and better real-time performance than traditional methods.