Hardware Acceleration for Neural Systems. To really understand how brains work, we need to simulate neural networks of a size similar to that of the human brain (100 billion neurons, 100 trillion connections). Simulating such a network on standard computers in not possible because of its sheer size. Several groups are currently building very expensive and proprietary hardware to solve this, but the output from these projects will not be accessible to other researchers. In order to make real prog ....Hardware Acceleration for Neural Systems. To really understand how brains work, we need to simulate neural networks of a size similar to that of the human brain (100 billion neurons, 100 trillion connections). Simulating such a network on standard computers in not possible because of its sheer size. Several groups are currently building very expensive and proprietary hardware to solve this, but the output from these projects will not be accessible to other researchers. In order to make real progress in neuroscience, many more researchers need to be enabled to participate. To do this, the project will build a system from commercial hardware (FPGAs) that will cost only a few ten thousand dollars and it will make this design and software available for free. Read moreRead less
Advanced planning systems for vertically integrated supply chain management. This project will integrate various algorithms into an adaptive, dynamic and intelligent system that deals with the vertically integrated supply chains. The outcomes include publications in the quality outlets, generation of intellectual property, and dissemination of this research amongst the research and business communities.
Mechanisms of learning at the interface between perception and action. Using the latest in brain imaging and simulator technology, this project will advance understanding of how experience shapes the visual centres of our brain. It will also support partnerships with construction, mining and health services by developing real and virtual machine interfaces and tools to enhance the outcome of simulator-based training.
Microcantilevers for multifrequency atomic force microscopy. This project aims to design a microcantilever with high-performing sensors more sensitive and with better noise performance than the typical optical system used in commercial Atomic Force Microscopes (AFMs). The AFM, a nanotechnology instrument, uses a microcantilever (with an extremely shape probe) to interrogate a sample surface. It has made important discoveries in nanotechnology, life sciences, nanomachining, material science and d ....Microcantilevers for multifrequency atomic force microscopy. This project aims to design a microcantilever with high-performing sensors more sensitive and with better noise performance than the typical optical system used in commercial Atomic Force Microscopes (AFMs). The AFM, a nanotechnology instrument, uses a microcantilever (with an extremely shape probe) to interrogate a sample surface. It has made important discoveries in nanotechnology, life sciences, nanomachining, material science and data storage systems. Despite its success, the technique’s spatial resolution and quantitative measurements are limited. This project could lead to breakthrough technologies such as atomic force spectroscopy to study elastic modulus of nanostructures, and establish Australia's prominence in this emerging field.Read moreRead less
Discovery Early Career Researcher Award - Grant ID: DE160100850
Funder
Australian Research Council
Funding Amount
$330,000.00
Summary
Dynamic adaptive software configuration. The aim of this project is to use computational intelligence in software engineering processes to achieve self-optimising products. Many mobile device users bemoan inadequate battery life, and device size is largely determined by the battery. In addition, information and communications technology uses six per cent of the global electricity production. The expected outputs of this project will be packages of optimisation components that software engineers ....Dynamic adaptive software configuration. The aim of this project is to use computational intelligence in software engineering processes to achieve self-optimising products. Many mobile device users bemoan inadequate battery life, and device size is largely determined by the battery. In addition, information and communications technology uses six per cent of the global electricity production. The expected outputs of this project will be packages of optimisation components that software engineers can incorporate into next generation products: the products analyse their collected usage data, perform what-if analyses, and optimise their configurations accordingly for the next usage period. Hence, the products may respond faster, be more reliable, and consume less energy.Read moreRead less
Tracing nature's template: using statistical machine learning to evolve biocatalysts. In this project new computational methods will be developed to design nature-inspired, biological catalysts for industrial purposes. Such methods will enable catalysts to be designed that can improve the effectiveness and environmental footprint of drug development, agricultural and specialist chemical production and environmental remediation.
Improving transient performance for systems with multiple inputs/outputs. This project aims to develop and test new mathematical techniques for the improvement of transient performance in tracking control systems. The fundamental problem to be addressed will be the design of controllers to rapidly track constant and time varying target reference signals without overshooting or undershooting for multiple-input multiple-output systems/plants. These new methods aim to offer improved accuracy and sp ....Improving transient performance for systems with multiple inputs/outputs. This project aims to develop and test new mathematical techniques for the improvement of transient performance in tracking control systems. The fundamental problem to be addressed will be the design of controllers to rapidly track constant and time varying target reference signals without overshooting or undershooting for multiple-input multiple-output systems/plants. These new methods aim to offer improved accuracy and speed in many engineering applications.Read moreRead less
Need for Speed: Towards Controller Design Automation for Power Electronics. This project aims to address the need for advanced controller design automation tools for power electronics systems by advocating a novel design paradigm. The project expects to seek breakthroughs in the modelling and optimisation aspects of power electronics systems and generate new automation tools for existing and emerging power electronics applications. Expected outcome include significant reduction of controller dev ....Need for Speed: Towards Controller Design Automation for Power Electronics. This project aims to address the need for advanced controller design automation tools for power electronics systems by advocating a novel design paradigm. The project expects to seek breakthroughs in the modelling and optimisation aspects of power electronics systems and generate new automation tools for existing and emerging power electronics applications. Expected outcome include significant reduction of controller development cycle time and cost, minimisation of human oversight, and maximisation of system performance. Profound benefits include maintaining Australia’s leadership in a wide range of sectors such as renewable energy and electric vehicles demanding rapid development cycles and realisation of Australia’s zero-carbon vision. Read moreRead less
System identification of microstructure in the brain using magnetic resonance. Magnetic Resonance Imaging technologies will be exploited to probe the microstructure of the brain, using powerful Bayesian optimisation techniques and innovative uses of magnetic resonance. The project will in particular develop non-invasive imaging methods to quantify iron content in the brain, important for research on dementia and Alzheimer's disease.
Reliable and efficient algorithms for modelling dynamical systems from data. Mathematical and computational models are increasingly important in diverse areas of science and engineering including aircraft and automotive design, robotics, medical sensing, and biology. However, finding an accurate model remains a difficult task. This project will develop new methods to reliably find highly accurate models from recorded data.