基于多分辨率分析的硅微陀螺随机漂移时间序列建模  

Time-Series Model of Silicon Micro-Gyroscope Random Drift Based on Multi-Resolution Analysis

作　　者：满海鸥 龙书林 肖定邦[2] 吴学忠[2] 王浩旭[2] 

机构地区：[1]总装驻常德地区军代室,湖南常德415007 [2]国防科技大学机电工程与自动化学院,长沙410073

出　　处：《传感技术学报》2012年第8期1107-1111,共5页Chinese Journal of Sensors and Actuators

摘　　要：针对硅微陀螺的随机漂移误差,根据多尺度分析理论,提出了随机漂移趋势项提取算法,并应用于时间序列分析,采用波克斯-詹金斯法建立了ARMA模型。进一步采用长自回归-白噪化建模方法对模型进行了辨识和适用性检验。最后,构造Kalman滤波器对ARMA模型进行了滤波,滤波后方差减小了一个数量级,硅微陀螺原始漂移的零偏稳定性为34.428°/h,Kalman滤波后零偏稳定性为2.34°/h,有效地提高了陀螺的使用精度。According to multi-resolution analysis theory,the arithmetic of gyroscope random drift trend is put forward which is applied to time-series analysis objected to the random drift error.Then,the Autoregressive Moving-Average model is set up by Box-Jenkins method.Further more,model identification and the validity test is taken by long autoregressive-white noise modeling method.Lastly,construct the Kalman filter to filtrate the ARMA model.The variance reduce one quantity order and the gyroscope bias offset stability improved from34.428 °/h to 2.34 °/h after Kalman filter,so its application precision can be further improved in practical system.

关 键 词：多分辨率分析 时间序列 ARMA模型 KALMAN滤波 

分 类 号：O224[理学—运筹学与控制论] V241.5[理学—数学]