Unknown Input Filtering of Linear Time-Varying Systems: A Parameterized Unknown Input Model Approach

摘要

It is well known for model-free unknown input filtering (MUIF) problem, the unknown inputs may be reconstructed with a multi-step delay, which may incur some serious filter implementation problems, such as time-delay smoothed estimates and the existence of unbiasedness condition. In order to alleviate the problems, this paper proposes a parameterized unknown input model (PUIM) through which an equivalent PUIM based system transformation is formed. It is shown that, within this new parameterized input filtering approach, the time-delay and filter existence condition problems encountered in the recently developed optimal simultaneous input and state estimation (OSISE) method can be solved. Two illustrative examples are given to show the effectiveness of the proposed results.