Advanced \(\mathcal H_\infty\) control. Towards nonsmooth theory and applications (Q390785)

Since in many control systems there is unmodeled dynamics, there is a natural need for controllers that don't only work well for the plant they are designed for, but also for plants in a neighborhood of this nominal plant. Controllers that achieve this are called robust. Since the distance between two linear models is measured with respect to the norm of the Hardy space \(H_{\infty}\), robust- and \(H_{\infty}\) -control have become almost synonyms. For linear, time-invariant systems the theory is well-developed and can now be found in (text)-books, \textit{B. A. Francis} [A course in \(H_{\infty}\) control theory. Lecture Notes in Control and Information Sciences, 88. Berlin etc.: Springer-Verlag. (1987; Zbl 0624.93003)], \textit{K. Zhou} and \textit{J. C. Doyle} [Essentials of robust control. Hemel Hempstead: Prentice Hall (1997; Zbl 0890.93003)] and Zhou, Doyle and Glover [\textit{K. Zhou} et al., Robust and optimal control. Upper Saddle River, NJ: Prentice Hall (1996; Zbl 0999.49500)]. As may be clear from the above, the robust design problem is equally relevant for a broader class of models. This is the aim of this book. The most interesting part of the book is part 3, in which the authors apply the results derived in the previous parts to practical examples. This is non-trivial, since although it may be mathematically clear how to extend the theory developed for linear time-invariant systems, it is far less obvious that the obtained results provide you with tools to actually calculate these robust controllers. As the authors indicate the present book is a research monograph. However, it is self-contained, and so the reader who takes the effort to study the book in detail, learns how to design robust controllers for a very general class of systems, such as those described by nonsmooth nonlinear models and those described by partial differential equations.