Reduced Order Kalman Filtering without Model ReductionControl & Intelligent Systems
AbstractThis paper presents all optimal discrete time reduced order Kalman filter. The reduced order filter is used to estimate a linear combination of a subset of the state vector. Most previous approaches to reduced order filtering rely on a reduction of the model order. However, this paper takes the full model order into account. The reduced order filter is obtained by minimizing the trace of the estimation error covariance.
Citation InformationSimon, D. D. (2007). Reduced Order Kalman Filtering without Model Reduction. Control & Intelligent Systems, 35(2), 169-174.