A state prediction for discrete nonlinear stochastic systems with a fixed advance

Аuthors

Pokalov V. A., Rudenko E. A.

Moscow Aviation Institute (National Research University), 4, Volokolamskoe shosse, Moscow, А-80, GSP-3, 125993, Russia

Abstract

A state prediction problem is considered for nonlinear stochastic systems with a fixed advance. An algorithm is derived to syntheses some finite-dimensional predictor with optimal structure. Gaussian approximation for a distribution function is used to construct predictor algorithm and its computer-based realization. A comparative precision analysis is performed for the Gaussian optimal-structure predictor and for the Gaussian absolutely optimal predictor based on some model problem.