A receding horizon kalman FIR filter is suggested for discrete time-varying systems,combining the Kalman filter with the receding horizon strategy.THe suggested filter is obtained by assuming that the horizon initial state is unknown.HTe filter is shown to be an FIR structure that has many good inherent porcperies.It can always be obtained irrespective of singularity problems caused by unknown informaiton about the horizon initial state.TH suggested filter is represtneed in an iterative form and also in a standard FIR form.THe suggesteedd filter turns out to be a remarkable deadbeat observer tha tis often robust against system and measurement noises.For the amenability to parallel and systolic implementation as well as the numericla stabnilty,a square-root algorithm for the suggested filter presented.
展开▼
机译：建议用于离散的时变系统的后退Horizo n Kalman FIR滤波器，将卡尔曼滤波器与后退地平线策略组合。通过假设地平线初始状态未知，可以获得建议的滤波器。HTE滤波器被显示为FIR结构有许多良好的固有豪猪。可以始终获得，而不管由关于地平线初始状态的未知信息引起的奇点问题，那么建议的过滤器是以迭代形式偿还的，也以标准的冷杉形式予以竞争。建议的滤波器变成了显着的Deadeat Observer Tha TIS通常稳健地防止系统和测量噪声。适用于并行和收缩的实施以及NumericLA稳定性，所提出的建议过滤器的平方根算法。
展开▼
