References
Even when restricted to control systems, the concept of robustness is quite broad and can be approached from many different angles. In our course we are restricting the focus to the approaches formulated in frequency domain. The main reference for this part of the course is the book (Skogestad and Postlethwaite 2005). The concepts and techniques introduced in our lecture are covered in Chapters 7 and 8 (up to 8.5) of the book.
What we typically do not cover in the lecture, but only due to time constraints, is the topic of structured uncertainties and their analysis using structured singular value (SSV, 𝜇). These are treated in the section 8.6 through 8.11 of the book. It is warmly recommended to have a look at it.
Although the book is not freely available online (only the first three pages are downloadable on the authors’ web page), it is available in a decent number of copies in the university library.
In case you are interested in getting the book in one way or another (perhaps even by purchasing it), make sure you get the second edition published 2005. The book contains some useful snippets of Matlab code and the first edition relies on some ancient version of Matlab toolboxes, which makes it useless these days.
The topic of modeling uncertainty in frequency domain using weighting filters plugged into additive or multiplicative structures is fairly classical now and as such can be found in numerous textbooks on robust control such as (Doyle, Francis, and Tannenbaum 2009), (Zhou, Doyle, and Glover 1995), (Dullerud and Paganini 2000), (Sánchez-Peña and Sznaier 1998). Although these are fine texts, frankly speaking they offer nearly no guidance for applying the highly advanced concepts to practical problems – they mostly focus on building up the theoretical framework. In this regard, Skogestad’s book is truly unique.
There are some newer texts that appear to be a bit more on the application side such as (Lavretsky and Wise 2024), (Yedavalli 2014) or (Gu, Petkov, and Konstantinov 2013), but I am not familiar them, to be honest.
Back to top