State Estimation of Linear Systems with State Equality Constraints
| Authors: | Ko Sangho, University of California, San Diego, United States
Bitmead Robert R., University of California, San Diego, United States |
|---|
| Topic: | 2.4 Optimal Control |
|---|
| Session: | Optimal Control and Estimation in Linear Systems |
|---|
| Keywords: | Estimation, Constraints, Kalman filters, Projection, Descriptor systems |
|---|
Abstract
This paper deals with the state estimation problem for linearsystems with state equality constraints. Using noisy measurementswhich are available from the observable system, we construct theoptimal estimate which also satisfies linear equality constraints.For this purpose, after reviewing modeling problems in linearstochastic systems with state equality constraints, we formulate aprojected system representation. By using the constrained Kalmanpredictor for the projected system and comparing its predictorRicccati Equation with those of the unconstrained and theprojected Kalman predictors, we reach the conclusion that thecurrent constrained estimator outperforms other filters forestimating the constrained system.
