November 10, 2010, 2001 ESB

In recent years, we have seen a growing interest in probabilistic and randomized methods for design of uncertain complex systems. In this lecture, we provide a broad perspective of this research area and discuss several randomized algorithms. In particular, we present sequential algorithms for convex problems and we discuss non-sequential algorithms for non-convex problems. Furthermore, we demonstrate the utility of this approach for analyzing specific applications, such as the computation of PageRank in search engine Google and the design of autonomous unmanned aerial vehicles.

I am interested in a broad range of feedback and dynamic systems theory and applications. Some of my recent interests and results are below.
I have a long standing interest in trying to analyses and understand the factors inherent in a system that limit the achievable feedback control performance, and hence gives rise to inevitable trade-offs in control system design. Early work on this examined the roles of open loop instability, non-minimum phase behaviour (that is, lack of a stable inverse), and time delays. More recently I have looked at communication constraints, and also problems in formation control of autonomous vehicles. In particular, I have worked on conditions in rigid formation following that give rise to a problem known as 'String Instability'.
For several years now, I have been part of the University of Newcastle's RoboCup soccer team, the NUbots. We have competed in the 4 Legged League Soccer, and have been placed in the top three since we joined the competition in 2002. In this application, I am particularly interested in the problem of robot localisation using the Extended Kalman Filter, feedback control for legged motion, and distributed systems. At NUIM, we have been accepted to compete in the standard platform league of the RoboCup soccer competition. Newcastle University (Australia) and NUI Maynooth competed as the NUManoids in the 2008 RoboCup Soccer, Standard Platform (Nao) League and achieved first place!
I have a wide range of experience in applications of systems, control and optimisation. A selection of recent work includes:
•Advanced Software Tools for Model Based Control (with Matrikon Pty Ltd)
•Scheduling and Stochastic Optimisation for Open Pit Mining Operations (with BHP Billiton Innovation)
•Automotive Control Applications
•Control of Power Electronics Inverters
I am interested in applications of systems, control and dynamics in biology, working with the Systems Biology group at the Hamilton Institute. Previous work has focussed on phase analysis of coupled oscillators. More recently I have been working with colleagues in the Systems Biology group at the Hamilton, in particular relating to understanding the dynamics of networks of biochemical reactions. I am also interested in modelling and control of HIV infection dynamics and Parkinson's disease.
have examined a number of aspects of feedback control loops over finite capacity communication channels. These works include examining information theoretic analyses of limits on the achievable loop properties dictated by finite capacity. In addition, I have worked on achievable performance for particular classes of power constrained AWGN and ACGN channels.