For maneuvering target tracking problem, robust filtering is an effective way to gain fast and accurate target trajectory in real time. The H ∞ filter (H ∞ F) is a conservative solution with infinite-horizon robustness, leading to excessive cost of filtering optimality and reduction of estimation precision. In order to retrieve the filtering optimality sacrificed by conservativeness of the H ∞ F design, in this paper, an optimal-switched filtering mechanism is developed and established on the standard H ∞ F to propose an optimal-switched H ∞ filter (OSH ∞ F). The optimal-switched mechanism adopts a switched structure that switches filtering mode between optimal and H ∞ robust by setting a switching threshold, and introduces an optimality-robustness cost function (ORCF) to on-line optimize the threshold such that the switching structure can be optimized. In the ORCF, a non-dimensional weight factor (WF) is used to quantify the ratio of the filtering robustness and optimality. As the only tunable parameter in the filter, when the WF is given, the proposed OSH ∞ F can obtain the optimal state estimates with filtering optimality and robustness kept at the WF-determined ratio. With the conservativeness of the H ∞ F optimized, the developed OSH ∞ F can be used as a generalized H ∞ F form. A simulation example of space target tracking has demonstrated the superior estimation performance of the OSH ∞ F compared with that of Kalman filter and other typical H ∞ filters.
展开▼
