自校正解耦信息融合Wiener状态估值器-控制理论与应用.PDF

25 4 Vol. 25 No. 4 2008 8 Control Theory Applications Aug. 2008 :2008 Wiener , ( , 150080) : , , (MA) , , , Wiener. Wiener. WienerWiener. , MA , Wiener, . 3. : ; ; ; ; Wiener : O211.64 : A Self-tuning decoupled information fusion Wiener state estimators DENG Zi-li, LI Chun-bo (Department of Automation, Heilongjiang University, Harbin Heilongjiang 150080, China) Abstract: For the multisensor systems with unknown noise variance matrices, using the modern time series analysis method, based on on-line identification of the moving average (MA) innovation model and the solution of the matrix equations for correlation function, the estimators of noise variance matrices can be obtained. Under the linear minimum variance optimal information fusion criterion weighted by scalars for components, a self-tuning decoupled information fusion Wiener state estimator is presented. Its accuracy is higher than each local self-tuning Wiener state estimator. It realizes the decoupled local Wiener estimators and decoupled fusion Wiener estimators for the state components. Its convergence is proved, i.e. if the parameter estimation of MA innovation model is consistent, then it will converge to the optimal decoupled information fusion Wiener state estimator with known noise variance matrices, so that it has asymptotic optimality. A simulation example for a target-tracking system with 3 sensors shows its effectiveness. Key words: multisensor information fusion; decoupled fusion; identification; noise variance estimation; self-tuning Wiener state estimator 1 (Introduction) Ka