Kalman 滤波自六十年代提出以来,由于其优良的性能而引起人们的关注,但经典的卡尔曼滤波算法需要系统的先验知识,并且计算量极大,通常只能用于数据的事后处理,使其应用受到限制.随着数据实时处理技术的发展,越来越多的人们转而寻求具有自适应能力的卡尔曼滤波算法,于是出现了一系列自适应算法,这些算法大都是为某种专用的目的而设计的,因而其自适应能力有限.其中以 Hampfon 的随机近似算法性能较好,但计算量较大. Since the 1960's, Kalman filter has attracted people's attention due to its excellent performance. However, the classical Kalman filter algorithm requires prior knowledge of the system and has a huge amount of computation, and can only be used for data post-processing. So its application is limited.With the development of data real-time processing technology, more and more people turn to seek adaptive Kalman filtering algorithm, so there has been a series of adaptive algorithms, these algorithms are mostly for some Purpose designed, so its adaptive ability is limited, among them Hampfon's stochastic approximation algorithm performs well, but its computation is large.