Towards a unified theory of constrained control and estimation. The project will investigate the implications of duality and other connections between constrained control and estimation. We believe that the research will result in a richer understanding of these problems. In particular, we envisage an impact in at least four areas: (i) Computational issues, i.e., development of more efficient algorithms for constrained problems. (ii) Geometry of constrained problems, by extending recent results ....Towards a unified theory of constrained control and estimation. The project will investigate the implications of duality and other connections between constrained control and estimation. We believe that the research will result in a richer understanding of these problems. In particular, we envisage an impact in at least four areas: (i) Computational issues, i.e., development of more efficient algorithms for constrained problems. (ii) Geometry of constrained problems, by extending recent results pertaining to constrained control to estimation problems. (iii) Problems with mixed constraints, for example, interval and finite set constraints. (iv) Fundamental limitations imposed by constraints to filtering and control problems.Read moreRead less
New System Identification Techniques Utilising Misspecified Models. National benefits of the proposed research project will result from improvements in control due to a better, more complete understanding of the models obtained by the newly proposed system identification technique. The resulting effect on industrial practice will be an increase in efficiency, by reduced waste, lower pollution levels and increased throughput. Also, the techniques developed will be directly applicable to current r ....New System Identification Techniques Utilising Misspecified Models. National benefits of the proposed research project will result from improvements in control due to a better, more complete understanding of the models obtained by the newly proposed system identification technique. The resulting effect on industrial practice will be an increase in efficiency, by reduced waste, lower pollution levels and increased throughput. Also, the techniques developed will be directly applicable to current research in the areas of complex systems, such as smart structures and biological studies of the dynamic effects of drugs and hormones on genes.Read moreRead less
Robust Experiment Design for Dynamical System Identification. Innovative and new robust experiment design methodologies are a Frontier Technology for Transforming Australian Industries. By providing a solid foundation for generating high fidelity models, robust experiment design will, by the use of breakthrough science, facilitate the estimation of models in minimum time. Also, this will entail minimal disruption to the normal operation of the process under study. With the majority of advanced ....Robust Experiment Design for Dynamical System Identification. Innovative and new robust experiment design methodologies are a Frontier Technology for Transforming Australian Industries. By providing a solid foundation for generating high fidelity models, robust experiment design will, by the use of breakthrough science, facilitate the estimation of models in minimum time. Also, this will entail minimal disruption to the normal operation of the process under study. With the majority of advanced industrial process control systems reliant on accurate models significant savings could also be made due to the implicit improvement in process control.Read moreRead less
Parsimonious Quantization in Signal Processing and Control. In today's society there is an abundance of data. Indeed, it could be argued that we suffer from data 'overload'. Thus to turn 'data' into actions, the need for parsimony in signal processing and control arises. For that purpose, the data must be sampled (in time) and quantized (in space). Within this context, the current project is aimed at understanding aspects of sampled parsimonious quantization. The results have widespread practica ....Parsimonious Quantization in Signal Processing and Control. In today's society there is an abundance of data. Indeed, it could be argued that we suffer from data 'overload'. Thus to turn 'data' into actions, the need for parsimony in signal processing and control arises. For that purpose, the data must be sampled (in time) and quantized (in space). Within this context, the current project is aimed at understanding aspects of sampled parsimonious quantization. The results have widespread practical uses including digital cameras, video compression, audio quantization, control over communication networks, switching of electronic devices and many others.Read moreRead less
Robust Dynamical System Identification. Innovative robust system identification methods are a Frontier Technology for Transforming Australian Industries. Robust system identification will provide a technology for generating high fidelity models by the use of breakthrough science. With the majority of advanced industrial control systems reliant on accurate models significant savings could be made due to the implicit improvement in process control. Furthermore, system identification is a key enabl ....Robust Dynamical System Identification. Innovative robust system identification methods are a Frontier Technology for Transforming Australian Industries. Robust system identification will provide a technology for generating high fidelity models by the use of breakthrough science. With the majority of advanced industrial control systems reliant on accurate models significant savings could be made due to the implicit improvement in process control. Furthermore, system identification is a key enabling technology in most modern systems (e.g. in aerospace, manufacturing, mining, minerals processing and telecommunications) and is also important in the emerging areas of nanotechnology and systems biology.Read moreRead less