Linear system identification as curve fitting

MaRDI QIDQ4451705zbMATH OpenFDO

Authors Lennart Ljung

Publication date 1 March 2004

zbMATH Keywords

curve fittingsystem identification

Mathematics Subject Classification ID

Numerical smoothing, curve fitting (65D10) System identification (93B30)

Review text

Linear system identification methods are set into the perspective of classical curve fitting techniques. Global and local parametric methods as well as non-parametric ones are discussed. System identification is viewed as the estimation of the frequency function curve. The empirical transfer function estimate is taken as the ``observations of this curve. The role of the noise model in dynamic models is also displayed. For the entire collection see [Zbl 1005.00024].

Recommendations

• scientific article; zbMATH DE number 3997615
• scientific article; zbMATH DE number 440537
• Frequency-domain system identification using non-parametric noise models estimated from a small number of data sets
• scientific article; zbMATH DE number 3930863
• Nonparametric preprocessing in system identification: a powerful tool

Cited in

(4)

• Initial estimates for the dynamics of a Hammerstein system
• Non-parametric linear time-invariant extensions of non-invertible and backlash plant
• scientific article; zbMATH DE number 4008195 (Why is no real title available?)
• Nonlinear gray-box identification using local models applied to industrial robots

Describes a project that uses

Uses Software

• System Identification Toolbox

This page was built for publication: Linear system identification as curve fitting

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q4451705)