Automation of information processing for detection of rail forms anomalies

Based on the analysis of experimental data, the analytical models of the information processing from the meters of inertial-satellite system, fixing the anomalies of the form of the rails are suggested. The models are described by the nonstationary differential equations with unknown parameters. The obtained equations for estimation of the state vector of an integrated navigation system allowed to develop a method for identifying the parameters of the rails anomaly, free from the need to expand the dimension of the filter and the assumption of stationarity of the identifiable parameters and providing the improvement of accuracy of identification by the use of probabilistic criteria nonlinearly dependent on the a posteriori probability density of the state vector.