基于系统误差及其协方差阵拟合的抗差自适应滤波

Adaptively robust filter based on synthetically fitting systematic errors and covariance matrices

动态导航数据处理中基于系统误差及其协方差矩阵拟合的自适应滤波算法在一定程度上可以抵制系统误差或区域性系统误差对动态导航定位结果的影响,但无法抵制异常扰动的干扰。考虑到观测模型和动力学模型存在异常扰动时,整体控制状态噪声协方差阵的方法可以减弱扰动异常的影响,因此提出一种基于系统误差及其协方差矩阵拟合的抗差自适应滤波算法。算例结果证实,该算法在减弱系统误差影响的同时,还能有效地抵制观测异常和载体状态扰动异常对动态系统参数估值的影响。

Adaptive filter combined with synthetically estimating systematic errors and covariance matrices of systematic error can be used to resist the effect that is induced by the systematic errors or area systematic errors during the data of dynamic navigation is processed, but it is difficult to control the abnormal disturbance. The influence of the abnormal disturbance can be reduced by the adjusting the covariance matrix of state noises when the observation model and dynamics model exist abnormal disturbance, so an arithmetic of adaptive robust filter is developed in this research. The results of simulated calculation show that the adaptive robust filter that is developed in this research not only can reduce the influence of the systematic errors, but also can resist the influence that induced by abnormal disturbance.