Péter Gáspár, Zoltán Szabó, József Bokor

Discrete Feedback Systems 2.

Modern Control


Maximum Likelihood and Bayesian Estimation

In this section we consider models described by a possibly nonlinear function M(θ) that maps from θRd into the space RN of measurements, YN. We sometimes drop the index N, i.e., write just yRN. The prediction model for obtaining the measurements is y=M(θ)+ε, where εRN is the measurement noise. We will treat two types of estimators: the maximum likelihood estimator and the Bayesian estimator, and conclude the section with the Cramer-Rao inequality, which gives a lower bound on the covariance matrix.

Discrete Feedback Systems 2.

Tartalomjegyzék


Kiadó: Akadémiai Kiadó

Online megjelenés éve: 2019

ISBN: 978 963 454 373 2

The classical control theory and methods that we have been presented in the first volume are based on a simple input-output description of the plant, expressed as a transfer function, limiting the design to single-input single-output systems and allowing only limited control of the closed-loop behaviour when feedback control is used. Typically, the need to use modern linear control arises when working with models which are complex, multiple input multiple output, or when optimization of performance is a concern. Modern control theory revolves around the so-called state-space description. The state variable representation of dynamic systems is the basis of different and very direct approaches applicable to the analysis and design of a wide range of practical control problems. To complete the design workflow, finally some introduction into system identification theory is given.

Hivatkozás: https://mersz.hu/gaspar-szabo-bokor-discrete-feedback-systems-2//

BibTeXEndNoteMendeleyZotero

Kivonat
fullscreenclose
printsave