zbMATH — the first resource for mathematics

Convergence rate of least-squares identification and adaptive control for stochastic systems. (English) Zbl 0604.93061
The paper considers multivariable (MIMO) ARMAX models whose parameters are estimated by the so-called extended least-squares (ELS) method (also called approximate maximum likelihood method, or pseudo-linear regression method). Under the usual positive realness condition and some other mild conditions, some results on the convergence rates for the parameter estimation error are derived. These results are subsequently used to analyse the performance of an adaptive control system with quadratic cost and ELS identifier. The rate of convergence of the cost to its minimal value is established.
Reviewer: P.Stoica

93E12 Identification in stochastic control theory
62M10 Time series, auto-correlation, regression, etc. in statistics (GARCH)
93C40 Adaptive control/observation systems
93C05 Linear systems in control theory
93C35 Multivariable systems, multidimensional control systems
Full Text: DOI
[1] CHEN H. F., Scientia sin. 25 pp 771– (1982)
[2] CHEN H. R., I.E.E.E. Trans, autom. Control 30 pp 189– (1985) · Zbl 0555.93057 · doi:10.1109/TAC.1985.1103906
[3] CHEN H. F., Kexue Tongbao (Science Bulletin) (1985)
[4] GOODWIN G. C., SIAM J. Control Optim. 19 pp 829– (1981) · Zbl 0473.93075 · doi:10.1137/0319052
[5] KAILATH T., Linear Systems (1980)
[6] LAI , T. L. , and WEI , C. Z. , 1982 ,Ann. Statist., 10 , 154 .
[7] LJUNG L., I.E.E.E. Trans, autom. Control 22 pp 551– (1976) · Zbl 0362.93031 · doi:10.1109/TAC.1977.1101561
[8] MOORE J. B., Automatica 14 pp 505– (1978) · Zbl 0398.93056 · doi:10.1016/0005-1098(78)90010-9
[9] SOLO V., I.E.E.E. Trans, autom. Control 24 pp 958– (1979) · Zbl 0422.62083 · doi:10.1109/TAC.1979.1102183
This reference list is based on information provided by the publisher or from digital mathematics libraries. Its items are heuristically matched to zbMATH identifiers and may contain data conversion errors. It attempts to reflect the references listed in the original paper as accurately as possible without claiming the completeness or perfect precision of the matching.