Two-Stage Optimal Estimators for a Class of Differential/ Algebraic Systems. Application to the State Estimation of Dynamical Systems with Partially Perfect Measurement

Optimal two-stage estimators are designed for standard stochastic dynamical systems which state is subject to exact delayed or undelayed algebraic constraints. The two-stage estimators are built upon the decoupling of the measurement update and the algebraic constraints update procedures, which yield a simple and attractive implementation scheme. We show how the obtained results can be applied to the problem of state estimation of systems having partially perfect measurement.