Joint System Identification for Point Processes and Time-series. In various application areas such as neurophysiology, earthquake modeling, price spikes in electricity markets, the data of interest are point processes (aka sequences of events) or combinations of point processes and analog signals. To understand the underlying subject of interest we need to be able to extract the maximum information from these observation sequences. The current tools for doing this are very limited. This resear ....Joint System Identification for Point Processes and Time-series. In various application areas such as neurophysiology, earthquake modeling, price spikes in electricity markets, the data of interest are point processes (aka sequences of events) or combinations of point processes and analog signals. To understand the underlying subject of interest we need to be able to extract the maximum information from these observation sequences. The current tools for doing this are very limited. This research program will develop the complex signal processing and system methodology needed to create a suitable tool set.Read moreRead less
Discovery Early Career Researcher Award - Grant ID: DE120102388
Funder
Australian Research Council
Funding Amount
$375,000.00
Summary
From Bayesian filtering to smoothing and prediction for multiple object systems. This project will develop new and improved algorithms for tracking multiple targets, such as tanks, submarines or planes, using the state of the art in mathematical and computational design. These will enable more efficient and accurate technologies for defence related applications including intelligence, surveillance and reconnaissance.
Parameter estimation for multi-object systems. Parameter estimation in multi-object system is essential to the application of multi-object filtering to a wider range of practical problems with social and commercial benefits. This project develops the necessary parameter estimation techniques for complete 'plug-and-play' multi-object filtering solutions that facilitates widespread applications.
A Bayesian framework for frequency domain identification. The national and social benefits of the project will be reflected
through the application recognition of the research work in the various industry and research community; and also through our international collaboration. The national and social benefits are also delivered by producing specialized researchers and engineers in systems and control engineering. These people include the research students who will participate in and learn f ....A Bayesian framework for frequency domain identification. The national and social benefits of the project will be reflected
through the application recognition of the research work in the various industry and research community; and also through our international collaboration. The national and social benefits are also delivered by producing specialized researchers and engineers in systems and control engineering. These people include the research students who will participate in and learn from the proposed project.Read moreRead less
A stochastic geometric framework for Bayesian sensor array processing. This project develops a mathematical framework, and a new generation of techniques, for sensor array processing to address real-world problems with uncertainty in array parameters and number of signals. The outcomes will enhance the capability of sensors in many application areas including, radar, sonar, astronomy and wireless communications.
Estimation and Control of Noisy Riemannian Systems. Many application areas such as satellite control, computer vision, coordination of rigid bodies, require the estimation and control of systems subject to geometric constraints. Most current algorithms for doing this are deterministic and can fail catastrophically in the presence of noise. This project aims to provide:
(i) Methods for analysing and then redesigning deterministic algorithms to ensure stability in the presence of noise;
(ii) New ....Estimation and Control of Noisy Riemannian Systems. Many application areas such as satellite control, computer vision, coordination of rigid bodies, require the estimation and control of systems subject to geometric constraints. Most current algorithms for doing this are deterministic and can fail catastrophically in the presence of noise. This project aims to provide:
(i) Methods for analysing and then redesigning deterministic algorithms to ensure stability in the presence of noise;
(ii) New design methods that deal with noise in an optimal way;
(iii) Noise resistant methods for distributed consensus seeking systems and cooperative control systems.
The outcomes will advance and benefit spatio-temporal data analysis and coordination in areas such as transport, health and video-security.Read moreRead less
Point processes system identification under simultaneity. Neuroscientists study neuronal brain dynamics of mammals via recordings from scores of tiny electrodes. But analysing these experiments is a problem because current methods cannot handle the common case where neurons discharge simultaneously. This project aims to provide powerful new tools to overcome this bottleneck.
Riemannian System Identification. A growing number of applications such as satellite attitude estimation, pose estimation in computer vision and direction estimation in statistics require the study of random processes in Riemannian manifolds and Lie Groups. This project will provide: methods for the construction/ numerical simulation of such processes; methods of system identification and their asymptotic performance analysis; and, algorithms for process state estimation.
Modeling stochastic systems in Riemannian manifolds. This project aims to develop new statistical signal processing and control engineering algorithms and tools that will enable tracking of objects remotely on land and in space. A growing number of applications require the study of random processes in Riemannian manifolds, that is processes that evolve subject to a geometric constraint. This project aims to provide methods for the numerical simulation of such processes, methods of online and off ....Modeling stochastic systems in Riemannian manifolds. This project aims to develop new statistical signal processing and control engineering algorithms and tools that will enable tracking of objects remotely on land and in space. A growing number of applications require the study of random processes in Riemannian manifolds, that is processes that evolve subject to a geometric constraint. This project aims to provide methods for the numerical simulation of such processes, methods of online and offline system identification from data on such processes and asymptotic performance analysis; and algorithms for process state estimation that obeys the geometry. The outcomes will advance and benefit spatio-temporal data analysis in areas such as transport, health and video-security.Read moreRead less
Vector network system identification. This machine learning project aims to provide more reliable ways to identify the structure and function of dynamic networks from both continuous and discrete network data. The project will use all the data and create principled new metrics. This could enable early diagnosis of network faults across a range of applications for example in power systems or diseased human brains. It could also enable discovery of critical functional subnetworks affecting reliabl ....Vector network system identification. This machine learning project aims to provide more reliable ways to identify the structure and function of dynamic networks from both continuous and discrete network data. The project will use all the data and create principled new metrics. This could enable early diagnosis of network faults across a range of applications for example in power systems or diseased human brains. It could also enable discovery of critical functional subnetworks affecting reliable operation in large complex human systems (such as financial systems) or natural systems (such as gene regulatory networks).Read moreRead less