卡尔曼滤波系统和量测噪声自适应估计的关联性

Relevancy of the Adaptive Estimation for System Noise and Measurement Noise of Kalman Filtering

中文关键词:  卡尔曼滤波  系统噪声和量测噪声  自适应估计  数据扰动情形

英文关键词:Kalman filtering  system noise and measurement noise  adaptive estimation  data disturbance situation

中文摘要:

      卡尔曼滤波是惯导系统(INS)/GPS组合导航的主要算法之一，Sage-Husa算法是在卡尔曼滤波基础上，为减少系统噪声和量测噪声的不确定性对误差估计的影响而采用的自适应估计方法。对Sage-Husa算法提出了4条改进措施；并通过在3种数据扰动情形下的仿真计算发现，只对一类噪声做自适应估计更容易产生较大的偏差，对系统噪声和量测噪声两类噪声同时做自适应估计，其效果要优于只对一类噪声做自适应估计，把此现象定义为卡尔曼滤波的系统和量测噪声自适应估计的关联性。这个结果不同于一些文献的观点。此项研究对自适应卡尔曼滤波在INS/GPS组合导航的工程化应用有较高的实用价值。

英文摘要:

      Kalman filtering is one of the main INS/GPS integrated navigation algorithms,Sage-Husa algorithm is the method based on Kalman filtering for alleviating the influence on the uncertainty of system noise and measurement noise.Four improvement measures are brought forward for the Sage-Husa algorithm.Through simulation computation in the three data disturbance situations,it is found that making adaptive estimation for one kind of noise more likely brings larger deviation,and the effect of synchronously making adaptive estimation for system noise and measurement noise is better than making adaptive estimation for only one kind of noise,this phenomenon is defined as the relevancy of the adaptive estimation for Kalman filtering system noise and measurement noise.This result is different from the viewpoint of some papers.The study has higher practical value for engineering application of adaptive Kalman filtering in INS/GPS integrated navigation.