报告人：Erik Weyer（The University of Melbourne）
Models of dynamical systems are widely used in science and engineering, and the field of system identification deals with the problem of building models from observed data. There will always be uncertainty in models of a real world systems, and it is important to assess the uncertainty. A common way to characterize the uncertainty in the model parameters is to use confidence regions and such regions can be constructed from asymptotic system identification theory. However, in asymptotic theory it is assumed that the number of data points tends to infinity, but the number of data points is of course always finite in any practical application. Methods for constructing guaranteed non-asymptotic confidence regions include the Leave-out Sign-dominant Correlation Regions (LSCR) and Signed Perturbed Sums (SPS) methods. Errors-in-Variable systems are characterised by the fact that both the input and the output are observed in noise, and this makes the construction of non-asymptotic confidence region difficult for such model. In this seminar we will give a brief overview of the LSCR and SPS methods, and present extensions of them applicable to Errors-in-Variable systems.
Erik Weyer is a Professor in the Department of Electrical and Electronic Engineering at the University of Melbourne, Australia. He received the Siv. Ing. degree in 1988 and the Ph.D. in 1993, both from the Norwegian Institute of Technology. From 1994 to 1996 he was a Research Fellow at the University of Queensland, Australia, before he joined the University of Melbourne in 1997. He has held visiting positions at the University of Brescia, Italy, the Technical University of Vienna, Austria, and Politecnico di Milano, Italy. From 2010 to 2012 he was Associate Editor of IEEE Transactions of Automatic Control and, and he is currently Associate Editor of Automatica. His theoretical research interests are in the areas of system identification while his applied work is focused on modelling and control water resources systems. He has collaborated with industry for more than 20 years, and in 2014 he was awarded the IEEE Control Systems Technology Award for the development and implementation of controls for irrigation channels and water management.