Simultaneous state and parameter estimation in linear systems.
Item
-
Title
-
Simultaneous state and parameter estimation in linear systems.
-
Identifier
-
AAI8801708
-
identifier
-
8801708
-
Creator
-
Eliazov, Teymuraz.
-
Contributor
-
Adviser: Frederick Thau
-
Date
-
1987
-
Language
-
English
-
Publisher
-
City University of New York.
-
Subject
-
Engineering, Electronics and Electrical
-
Abstract
-
In this work we consider the simultaneous state and parameter estimation problem of a linear discrete-time system subject to an arbitrary known input, and random input (driving noise), and with noisy output observations. A simultaneous state and parameter estimation problem is formulated as a least squares minimization problem. The special structure of the least squares minimization problem, allows separation of the original problem into two subproblems that are explicitly coupled: a nonlinear functional minimization to obtain parameter estimates and a linear least-squares problem to obtain state estimates. A new, computationally economical and robust algorithms are derived based on the proposed method. A theoretical and experimental study of the asymptotic properties of the parameter and state estimates are presented.;An adaptive control system for fine pointing of a flexible spacecraft is designed. Derived algorithms are used to simultaneously identify unknown states and parameters of a discrete-time lumped-parameter model of the flexible structure. The identified states and parameters form the input to a bang-off-bang control law that, in conjunction with the identification algorithm, results in an adaptive system where response closely approximates that of a system with known parameters. Simulation studies demonstrate the response achievable with the proposed approach.
-
Type
-
dissertation
-
Source
-
PQT Legacy CUNY.xlsx
-
degree
-
Ph.D.
