Fast identification of paramters in optical systems (Q1273163)
From MaRDI portal
![]() | This is the item page for this Wikibase entity, intended for internal use and editing purposes. Please use this page instead for the normal view: Fast identification of paramters in optical systems |
scientific article
Language | Label | Description | Also known as |
---|---|---|---|
English | Fast identification of paramters in optical systems |
scientific article |
Statements
Fast identification of paramters in optical systems (English)
0 references
25 July 2000
0 references
This paper develops a scheme for fast identification of perturbed physical parameters in optical systems (precise astronomical telescope) from which large-order measurement vectors are available. There are two parameters to be identified, namely the vertical and horizontal perturbations in the position of the hyperbolic mirror. The identification scheme consists of two stages. The first stage is a computationally intensive off-line procedure that must be performed only once, usually with a simulation model. The second stage is a fast identification algorithm that can be run repeatedly in real time. The coefficient vectors in this scheme are determined by a least-square fit. The exploitable pseudo-inverse matrix is computed only once, offline, and stored for use in the second stage of the identification scheme. The two-dimensional optical model is introduced. Solving the ray-trace equations yields the optical path-length difference between the length of the rays for the nominal wave front and the length of the rays for the wave front with aberration. The state space equation for the mirror dynamics is the system of linear differential equations with white Gaussian noise. The estimate of each parameter vector is the input to a linear-quadratic-Gaussian feedback controller. A discrete stationary Kalman filter generates a state estimate. A linear-quadratic regulator determines a linear feedback control to minimize the quadratic design criterion. The parameters of optimal feedback control law are the solution of the Riccati equation. The feedback control loop suppresses vibrations in the antenna. Several numerical results of identification and feedback control of perturbations in the position of the hyperbolic mirror in the two-mirror antenna are presented.
0 references
parameter identification
0 references
feedback control
0 references
real-time identification and control
0 references
optical systems
0 references
space antennas
0 references