The comparison of the Monte-Carlo method and neural networks algorithms in nonlinear estimation prob

Abstract The paper compares the algorithms based on neural networks and the Monte-Carlo method as applied to nonlinear estimation problems solved in the framework of the Bayesian approach. Two variants are considered. The first variant is a search of optimal estimates that are conditional mathematical expectations and, in a general case, depend on measurements in a nonlinear way. The second variant involves linear optimal estimates. In designing them, the root-mean-square criterion is minimized in the class of estimates that are linearly dependent on measurements. The comparison results are discussed.

[1]  Oleg S. Amosov,et al.  Linear optimal estimation algorithms based on the Monte Carlo method and neural networks for nonlinear navigational problems , 2006, 2006 IEEE Conference on Computer Aided Control System Design, 2006 IEEE International Conference on Control Applications, 2006 IEEE International Symposium on Intelligent Control.

[2]  James S. Meditch,et al.  Stochastic Optimal Linear Estimation and Control , 1969 .

[3]  F. Daum.  Nonlinear filters: beyond the Kalman filter , 2005, IEEE Aerospace and Electronic Systems Magazine.

[4]  A. Jazwinski.  Stochastic Processes and Filtering Theory , 1970 .

[5]  Niclas Bergman,et al.  Recursive Bayesian Estimation : Navigation and Tracking Applications , 1999 .

[6]  H. Sorenson,et al.  Stochastic optimal linear estimation and control , 1972, IEEE Transactions on Automatic Control.

[7]  James Chalfee,et al.  Bancroft's Algorithm is Global Nonlinear Least Squares , 1996 .