Recursive state estimators for general discrete-time stochastic nonlinear models

The design of recursive state estimators is considered for a general class of nonlinear stochastic discrete-time models. By deriving equations for a bound on the estimation error covariance based on a fixed-structure estimator, “bound-optimal” finite time minimum variance estimator parameters are found. The existence conditions for the statistical steady state are derived and applied to the design of constant gain estimators. A simulation result illustrates the use of such estimators