Title :
On optimal LTI approximation of nonlinear systems
Author :
Mäkilä, Pertti M.
Author_Institution :
Autom. & Control Inst., Tampere Univ. of Technol., Finland
fDate :
7/1/2004 12:00:00 AM
Abstract :
Optimal linear time-invariant (LTI) approximation of discrete-time nonlinear systems is studied using a quasi-stationary signal description. The optimality of the LTI approximation is quantified as a least average squared error condition, reminescent of recent studies on the performance of least squares type identification methods for LTI models of nonlinear systems.
Keywords :
discrete time systems; least squares approximations; nonlinear control systems; optimal control; discrete time nonlinear system; least squares methods; optimal linear time-invariant approximation; quasi stationary signal description; Control theory; Least squares approximation; Least squares methods; Linear approximation; Nonlinear control systems; Nonlinear distortion; Nonlinear systems; Space technology; Terminology; Time varying systems; Approximation; least squares; linear models; nonlinear systems; quasi-stationarity;
Journal_Title :
Automatic Control, IEEE Transactions on
DOI :
10.1109/TAC.2004.831120
