Improving Linear State-Space Models with Additional Iterations
Paper in proceeding, 2018
An estimated state-space model can possibly be improved by further iterations with estimation data. This contribution specifically studies if models obtained by subspace estimation can be improved by subsequent re-estimation of the B, C, and D matrices (which involves linear estimation problems). Several tests are performed, which show that it is generally advisable to do such further re-estimation steps using the maximum likelihood criterion. Stated more succinctly in terms of MATLAB® functions, ssest generally outperforms n4sid.
subspace identification
parameter estimation
maximum likelihood
state-space models
Author
Suat Gumussoy
MathWorks Inc
Ahmet Azda Ozdemir
MathWorks Inc
Tomas McKelvey
Chalmers, Electrical Engineering, Signal Processing and Biomedical Engineering
Lennart Ljung
Linköping University
Mladen Gibanica
Chalmers, Mechanics and Maritime Sciences (M2), Dynamics
Volvo Cars
Rajiv Singh
MathWorks Inc
IFAC-PapersOnLine
24058971 (ISSN) 24058963 (eISSN)
Vol. 51 15 341-346
18th IFAC Symposium on System Identification SYSID 2018
Stockholm, Sweden,
Stockholm, Sweden,
Subject Categories (SSIF 2011)
Control Engineering
Signal Processing
DOI
10.1016/j.ifacol.2018.09.158