论文信息 - A discussion of "unknown-but-bounded" disturbances in system identification

A discussion of "unknown-but-bounded" disturbances in system identification

In this paper, we point out that a fundamental property of a disturbance is that it is independent of the input-otherwise it is rather part of the system. Based on this characterization we show that parameter convergence can be obtained not only for stochastic but also for unknown-but-bounded disturbances if the input is at our disposal.<<ETX>>

H. Hjalmarsson | L. Ljung | H. Hjalmarsson | L. Ljung

[1] Petre Stoica,et al. Decentralized Control , 2018, The Control Systems Handbook.

[2] A. Kurzhanski. Identification - a theory of guaranteed estimates , 1989 .

[3] L. Ljung,et al. Hard frequency-domain model error bounds from least-squares like identification techniques , 1992 .

[4] Y. F. Huang,et al. On the value of information in system identification - Bounded noise case , 1982, Autom..

[5] Fred C. Schweppe,et al. Uncertain dynamic systems , 1973 .

[6] R. Tempo,et al. Optimal algorithms theory for robust estimation and prediction , 1985 .

[7] J. Deller. Set membership identification in digital signal processing , 1989, IEEE ASSP Magazine.

[8] E. Walter,et al. Exact and recursive description of the feasible parameter set for bounded error models , 1987, 26th IEEE Conference on Decision and Control.