Font Size: a A A

Filter, Based On The Arma Innovation Model, Self-tuning Information Fusion

Posted on:2010-09-15Degree:MasterType:Thesis
Country:ChinaCandidate:W L WangFull Text:PDF
GTID:2208360275992714Subject:Control theory and control engineering
Abstract/Summary:PDF Full Text Request
As a new technology of multisource data processing, multisensor information fusion estimation detects the same object by multisensors, avoids the limitation of single sensor, so it can offer more general and accurate information,further it can synthesize multisource data from the object, generate more accurate estimation than single source.For the multisensor systems with correlated measurement noises and unknown model parameters and noise statistics, using the modern time series analysis method, based on the Autoregressive Moving Average(ARMA) innovation models, a multi-stage identification algorithm is presented, where the on-line model parameters estimators can be obtained by recursive extended least squares method(RELS) or Gevers-Wouters method(G-W), and based on the solution of the matrix equations for correlation function,the on-line noise statistics estimators can be obtained. For the multisensor systems with unknown noise statistics, under the linear minimum variance optimal information fusion criterion weighted by scales for components, the self-tuning information fusion Kalman and Wiener filters and predictors are presented respectively. For the single channel ARMA signal with unknown model parameters and noise statistics, a self-tuning information fusion Wiener filter is presented.The Dynamic Error System Analysis(DESA) method for convergence analysis is extended, where the two decision criterions of the stability of Lyapunov equation are presented and proved, and then the convergence of the above mentioned self-tuning fusion estimators are proved, that is,the self-tuning fusers converge to the corresponding steady-state optimal fusers with probability one.Many simulation examples of tracking system show their effectiveness.
Keywords/Search Tags:multisensor information fusion, modern time series analysis method, self-tuning information fusion Kalman filter and Wiener filter, convergence
PDF Full Text Request
Related items