论文部分内容阅读
卡尔曼滤波(KF)广泛应用于线性系统的状态估计问题.然而,它需要精确已知过程噪声的统计特性,这在实际应用中往往是不能满足的.在这个背景下,首先根据协方差匹配原理建立了一种带有过程噪声递推估计的自适应KF算法.随后,为了突破KF只能处理线性系统估计问题的局限,将过程噪声递推估计引入到集合卡尔曼滤波(En KF)中,提出了一种自适应En KF算法.最后采用估计理论证明了新算法的稳定性.与标准Kn KF相比,该自适应算法在过程噪声统计特性未知的情况下滤波依然收敛,滤波精度及稳定性显著提升.大量仿真结果验证了所提出算法的有效性.
Kalman filter (KF) is widely used in the state estimation of linear systems, however, it needs to accurately know the statistical characteristics of process noise, which is often not satisfied in practical applications.In this context, firstly, according to covariance matching Principle to establish an adaptive KF algorithm with process noise recursion estimation.Then, to overcome the limitation that KF can only deal with the problem of linear system estimation, process noise recursion estimation is introduced into ensemble Kalman filter (En KF) , An adaptive En KF algorithm is proposed.Finally, the stability of the new algorithm is proved by using the estimation theory.Compared with the standard Kn KF, the filtering algorithm still converges when the statistical characteristics of the process noise are unknown, the filtering accuracy and The stability is improved significantly.A large number of simulation results verify the effectiveness of the proposed algorithm.