Modeling and identifying of Fiber Optic Gyro Random Drift Data Based on Time Series Analysis Method
-
摘要: 为了对光纤陀螺(FOG)随机漂移(启动漂移、慢变漂移和快变漂移)的各项参数进行快速有效的辨识，建立了含未知模型参数(慢变漂移的相关系数)的FOG随机漂移状态方程模型，通过对FOG随机漂移进行时间序列分析，辨识了未知的模型参数以及慢变漂移和快变漂移的自噪声方差，以卡尔曼滤波/平滑技术计算了FOG随机漂移的3 项分量。对实测的FOG随机漂移数据进行了模型辨识与适用性检验，结果表明，该方法能够准确计算FOG随机漂移的各项参数和分量。通过建立FOG随机漂移的状态方程模型，首次将时间序列分析方法与卡尔曼滤波/平滑技术相结合，快速计算了启动漂移、慢变漂移的相关系数以及慢变漂移和快变漂移的自噪声方差，这对于修正FOG单独工作或与其他设备组合时的导航误差具有重要的参考价值。Abstract: In order to effectively and quickly identify the parameters of random drifts (startup, slow and fleet) of fiber optic gyro( FOG) , the random drift of FOG is modeled as a equation of state with unknown model parameters (correlation coefficient of slow drift) , the unknown model parameters and white noise covariance of slow and fleet drifts are identified based on the time series analysis , and the coupling among three items of the random drift is computed with Kalman filter/ smoother. The real data of random drift of FOG is identified with the model to verify its applicability , and the result suggests that this method accurately compute all of the random drifts and the parameters of FOG. The present method could be used for revising navigation error of FOG as it works alone or in combination with other equipment.
-
[1] 邓自立.自校正滤波理论及其应用[M].哈尔滨:哈尔滨工业大学出版社,2003. [2] 邓自立.最优估计理论及其应用[M]. 哈尔滨:哈尔滨工业大学出版社,2005. [3] 秦永元,张洪锁,汪叔华,等.卡尔曼滤波与组合导航原理[M].西安:西北工业大学出版社,1998. [4] 王振龙.时间序列分析[M].北京中国统计出版社,2002. [5] 李言俊,张科.系统辨识理论及应用[M].北京:国防工业出版社,2003. [6] 喊荣春,崔平远.陀螺随机漂移时间序列建模方法研究[J].系统仿真学报,2005,17(8):1845-1847. [7] 阙志宏.线性系统理论[M].西安:西北工业大学出版社,1994. -
点击查看大图
计量
- 文章访问数: 1958
- HTML全文浏览量: 0
- PDF下载量: 855
- 被引次数: 0
下载:
