
Accession Number : AD0743221
Title : Optimal MinimalOrder Observers for DiscreteTime SystemsA Unified Theory,
Corporate Author : CALIFORNIA UNIV LOS ANGELES SCHOOL OF ENGINEERING AND APPLIED SCIENCE
Personal Author(s) : Leondes,C. T. ; Novak,L. M.
Report Date : 18 MAY 1972
Pagination or Media Count : 10
Abstract : Luenberger's minimalorder observer is considered as an alternate to the Kalman filter for obtaining state estimates in linear discretetime stochastic systems. The general solution to the problem of constructing the optimal minimalorder observer is presented for systems having white noise disturbances. In the special case of no measurement noise the observer estimation errors are shown to be identical with those of the corresponding Kalman filter. Estimation errors comparable with the Kalman filter are obtained when measurement noise is not excessive. The observer solution is extended to systems for which the noise disturbances are timewise correlated processes of the Markov type. In considering correlated noise inputs, the system state equations are not augmented as is done in the usual Kalman filtering theory. The observer solution, modified appropriately to account for the timewise correlation of the noise inputs, yields minimum meansquare estimates of the state vector. (Author)
Descriptors : (*ADAPTIVE CONTROL SYSTEMS, MATHEMATICAL MODELS), LINEAR SYSTEMS, STOCHASTIC PROCESSES, WHITE NOISE, OPTIMIZATION
Subject Categories : Statistics and Probability
Distribution Statement : APPROVED FOR PUBLIC RELEASE