Description

Abstract:
Theory and techniques for the identification of dynamic models from experimentally obtained system input-output data.


Objective:
To provide a series of practical techniques for the development of dynamical models from experimental data, with the emphasis being on the development of models suitable for feedback control design purposes. To provide sufficient theory to enable the practitioner to understand the trade-offs between model accuracy, data quality and data quantity.

Content:
- Introduction to modeling: Black-box and grey-box models; Parametric and non-parametric models; ARX, ARMAX (etc.) models.

- Predictive, open-loop, black-box identification methods. Time and frequency domain methods. Subspace identification methods.

- Optimal experimental design, Cramer-Rao bounds, input signal design.

- Parametric identification methods. On-line and batch approaches.

- Closed-loop identification strategies. Trade-off between controller performance and information available for identification.

Prerequisites:
Control systems (227-0216-00L) or equivalent. Familiarity with the following concepts is assumed:
- Laplace and Fourier transforms;
- Z-transform;
- Differential and difference equations;
- State-space representations;
- Basic stochastic variable concepts.

General Information

Lectures:
Wednesdays 10.15-12.00 in HG E 1.2
Exercise session:
Wednesdays 12.15-13.00 in ETZ D 61
Office hours:
Fridays 14.00-15.00 in ETL I 10.1
Teaching assistants:
Eva Ahbe: ahbee(at)control.ee.ethz.ch
Paul Beuchat: beuchatp(at)control.ee.ethz.ch
Georgios Darivianakis: gdarivia(at)control.ee.ethz.ch
Benjamin Flamm: flammb(at)control.ee.ethz.ch
Tony Wood: woodt(at)control.ee.ethz.ch
Grading scheme:
A midterm project and a take home final examination. Both will involve submitting Matlab solutions via email. The midterm will be worth 35% and the final will be worth 65%.
Collaboration, cooperation and plagiarism
Student's are encouraged to discuss exercise problems amongst themselves and in particular to participate in answering questions in the Piazza forums.

However any work that is to be submitted for a grade (in particular the midterm project and the final examination) must be done individually without any discussion with anyone else. If you believe some aspect of the graded work is unclear or incorrect, please contact Prof. Smith via the Piazza site.

Announcements

Announcements are not public for this course.
Staff Office Hours
NameOffice Hours
Roy Smith
When?
Where?

Lecture Notes