MEMS陀螺随机误差建模与补偿  

Random error modeling and compensation for MEMS gyroscope

作　　者：吕印新[1] 肖前贵[2] 杨柳庆[2] 

机构地区：[1]南京航空航天大学自动化学院,南京210016 [2]南京航空航天大学无人机研究院,南京210016

出　　处：《电子测量技术》2012年第12期41-45,共5页Electronic Measurement Technology

摘　　要：MEMS陀螺精度较低,随机漂移较大,严重影响系统的性能。对MEMS陀螺随机误差进行了时间序列分析,并建立了ARMA模型。使用MATLAB计算所选模型参数建立随机误差的系统方程,采用经典卡尔曼滤波器验证了在静态条件下,滤波后的信号标准差为滤波前的3.88%。针对动态条件下,常规卡尔曼滤波器滤波效果下降的问题,推导并设计了渐消卡尔曼滤波器。仿真结果表明,渐消卡尔曼滤波器能显著改善动态条件下的滤波效果,并且滤波精度较高。MEMS gyroscopes usually have low precision and very large random drift which seriously affect on system performance. The paper analyzed the stochastic error by time series and set up ARMA model. We used MATLAB to compute the parameters and establish system equation of stochastic error . After filtering in the case of static, the standard deviation of error were 9. 06M of that before filtering.. In view of typical KF＇s bad performance under oscillating environment,a kind of adaptive fading Kalman filter was deduced and designed. The results showed that the adaptive fading Kalman filter could get a remarkable performance improvement.

关 键 词：随机误差 陀螺 渐消卡尔曼滤波 时间序列 

分 类 号：V241.6[航空宇航科学与技术—飞行器设计]