The University of Southampton

ECS PhD thesis on Control judged best in the UK by IET

Published: 16 June 2010
Illustration

Dr Dominic Buchstaller, Research Fellow in the School of Electronics and Computer Science, has received this year’s Control and Automation Prize from the Institution of Engineering and Technology (IET) for the best UK PhD thesis in the subject.

Dr Buchstaller carried out the research for his thesis – Robust Stability and Performance for Multiple Model Switched Adaptive Control in the Information: Signals, Images, Systems research group in ECS under the supervision of Dr Mark French.

“Dominic's thesis has made a profound contribution to the field of adaptive control,” said Mark. “Whilst seen by many to be the key to providing controllers that work in highly complex and uncertain environments, many of the algorithms developed have significant robustness problem: when transferred from the ideal mathematical world to the real systems, they often fail, sometimes catastrophically.

“Dominic's thesis provides an entirely fresh and novel approach to the problems of robust design and synthesis for a wide class of algorithms and represents an important breakthrough in solving these decades-old problems.”

Dominic commented: "I find adaptive control a very interesting area to work in, since one knows a priori that the concept of adaptation works: for example humans naturally use it to deal with control tasks that involve large uncertainties, i.e. the picking-up of a packing case of unknown weight (books or pillows?). Then you have a great deal of research results on the engineering counterparts of natural adaptive systems, the classical adaptive control algorithms, that are designed to deal with exactly these kind of large parametric uncertainties. Although these controllers work on paper they usually fail in real-world applications, revealing a disconnect between theory and practice.

"Fortunately, more recent (multiple model type) adaptive control algorithms work rather well in practice. However, as with their classical counterparts, they usually lack a coherent theoretical framework that tells us about their (robustness and performance) properties and how to use them in practice. This is exactly what I attempted to provide with my thesis."

Professor Harvey Rutt, Head of the School of Electronics and Computer Science, commented: “Dominic’s research characterizes two important strands of the School’s research work – our commitment to real-world applications and our innovative approach to fundamental problems. We are delighted with this very distinguished recognition for the quality and success of Dominic’s research.”

For further information contact Joyce Lewis; tel.(0)23 8059 5453.

Articles that may also interest you

Share this article FacebookTwitterWeibo