Recursive estimation of model error for identification of nonlinear continuous time systems

Raol, JR and Parameswaran, V (1993) Recursive estimation of model error for identification of nonlinear continuous time systems. In: Symposium on System identification, 24-25 Nov 1993, Bangalore, India.

Full text not available from this repository.

Abstract

In this paper an Invariant Embedding Based Model Error Estimation (IEBMEE) algorithm for estimation of deterministic discrepancy in the assumed model of continuous time nonlinear systems is presented. The IEBMEE algorithm processes data in recursive manner and has appealing features similar to extended Kalman filter. It explicitly provides the estimates of the model error, which is then parameterized via least squares method to obtain the coefficients of another model that would explain the deterministic deficiency in the chosen model to match the true model in the sense of minimum model error. The numerical simulation results are obtained by implementing the continuous time IEBMEE algorithm in PC MATLAB.

Item Type: Conference or Workshop Item (Paper)
Uncontrolled Keywords: Dynamical systems;nonlinear equations;Recursive functions; Parameter identification;Error analysis;Kalman filters;Boundary value problems;Least squares method
Subjects: AERONAUTICS > Aerodynamics
Division/Department: Flight Mechanics and Control Division, Flight Mechanics and Control Division
Depositing User: Mrs Manoranjitha M D
Date Deposited: 21 Dec 2006
Last Modified: 24 May 2010 09:54
URI: http://nal-ir.nal.res.in/id/eprint/3797

Actions (login required)

View Item