Development of a three dimensional audio-visual next generation speech recognition system. To overcome the disadvantages of current Audio-Visual Speech Recognition Systems, we propose a set of robust algorithms in three dimensional computer vision and speech processing. The proposed system will have far-reaching implications in various areas, for example, human-machine interaction for speech recognition in automated dialog systems and voice-to-text conversions.
A neural fuzzy fusion engine for human-machine autonomous systems. This project aims to develop an intelligent engine to adaptively fuse multiple trust-based information from various agents in human machine autonomous systems (HMAS). The project will develop new techniques to detect covert-state drift, model trustworthiness between humans and machines, and adaptively fuse information under various kinds of uncertainty and trust levels. These techniques will be integrated to produce a general fra ....A neural fuzzy fusion engine for human-machine autonomous systems. This project aims to develop an intelligent engine to adaptively fuse multiple trust-based information from various agents in human machine autonomous systems (HMAS). The project will develop new techniques to detect covert-state drift, model trustworthiness between humans and machines, and adaptively fuse information under various kinds of uncertainty and trust levels. These techniques will be integrated to produce a general framework to facilitate human-machine interaction and enable better collaborative decisions in HMAS. The outcomes will benefit human-centric automation systems in general and next-generation autonomous vehicles in particular, which will contribute to the Australian economy.Read moreRead less
Collision Avoidance in Shipping Lanes via Intelligent Sensor Data Fusion . This project aims to develop an online maritime traffic monitoring system for reliable collision/contact avoidance that exploits complementary data from high-resolution airborne sensors and surface vessel sensors. Our approach is based on optimal scheduling and fusion of the sensor data and possibly other sources of data to construct a comprehensive dynamic picture of maritime traffic, in real-time. Moreover, the proposed ....Collision Avoidance in Shipping Lanes via Intelligent Sensor Data Fusion . This project aims to develop an online maritime traffic monitoring system for reliable collision/contact avoidance that exploits complementary data from high-resolution airborne sensors and surface vessel sensors. Our approach is based on optimal scheduling and fusion of the sensor data and possibly other sources of data to construct a comprehensive dynamic picture of maritime traffic, in real-time. Moreover, the proposed methodology enables quantification of confidence in the predictions. This will provide ship owners, directly to their vessels and/or at the fleet management centres, information such as weather reports, reliable collision/no-collision warnings and avoidance strategies, on-the-fly. Read moreRead less
A coupled finite volume method for viscoelastic flow problems on highly-skewed unstructured meshes: a computational rheology revolution. Commercial tools are unavailable for 21st century industry to analyse complex flow processes involving viscoelastic materials. Using fabrication of microstructured polymer optical fibre as a key case study, a coupled finite volume methodology holds the key for the next generation of computational rheology simulators.
Deep Learning that Scales. Deep learning has dramatically improved the accuracy of a breathtaking variety of tasks in AI such as image understanding and natural language processing. This project addresses fundamental bottlenecks when attempting to develop deep learning applications at scale. First, this project proposes efficient neural architecture search that is orders of magnitude faster than previously reported, abstracting away the most complex part of deep learning. Second, we will desig ....Deep Learning that Scales. Deep learning has dramatically improved the accuracy of a breathtaking variety of tasks in AI such as image understanding and natural language processing. This project addresses fundamental bottlenecks when attempting to develop deep learning applications at scale. First, this project proposes efficient neural architecture search that is orders of magnitude faster than previously reported, abstracting away the most complex part of deep learning. Second, we will design very efficient binary networks, enabling large-scale deployment of deep learning to mobile devices. Thus this project will overcome two primary limitations of deep learning generally, however, and will greatly increase its already impressive domain of practical application.Read moreRead less
Implementation of cognitive radar techniques in resource limited radar systems. Cognitive radar technology enables a multiple functional radar system to be built on a single chip, to be of high efficiency and low cost. Waveform design and scheduling play a key role in such a system. This project will investigate and design waveforms and scheduling methods for building a real cognitive radar system in the extremely high frequency band.
A Bayesian Approach to Distributed Estimation for Multi-Object Systems. This project aims to develop new signal processing techniques that facilitate autonomous technologies for environmental perception, with the ability to efficiently process large data volumes from multiple sensing modalities. Rapid advances in sensors and networks have led to a digital data deluge, from which extracting useful information presents new technological challenges and opportunities. To address this development, th ....A Bayesian Approach to Distributed Estimation for Multi-Object Systems. This project aims to develop new signal processing techniques that facilitate autonomous technologies for environmental perception, with the ability to efficiently process large data volumes from multiple sensing modalities. Rapid advances in sensors and networks have led to a digital data deluge, from which extracting useful information presents new technological challenges and opportunities. To address this development, this project seeks to develop new distributed solutions for statistical estimation, which are specifically designed for dynamic systems with multiple object states, and are inherently scalable and robust. The potential benefits include new technologies for smart cities, autonomous infrastructure, and digital productivity.Read moreRead less
Shared-space interactions between people and autonomous vehicles. This project aims to understand how autonomous vehicles in urban environments need to interact with the people that they share those spaces with. Autonomous vehicles that are able to operate in shared spaces, such as campuses and pedestrian zones, promise to improve urban life. However, their uptake depends heavily on public acceptance as they operate in close proximity to people. The project investigates whether people are more l ....Shared-space interactions between people and autonomous vehicles. This project aims to understand how autonomous vehicles in urban environments need to interact with the people that they share those spaces with. Autonomous vehicles that are able to operate in shared spaces, such as campuses and pedestrian zones, promise to improve urban life. However, their uptake depends heavily on public acceptance as they operate in close proximity to people. The project investigates whether people are more likely to trust the technology and feel safe if they are able to understand how the system makes decisions and to directly influence its behaviour. Outcomes are expected to promote safe behaviour around urban robotic applications and accelerate the uptake of autonomous systems in Australia’s cities. Read moreRead less
Design guidelines for safety-critical controllers in high-risk environments. This project aims to generate novel product design guidelines for developing safer controllers for use by potentially stressed individuals in high-risk situations. It will do this by generating specific insights and verifying generalisable solutions from the context of total artificial heart recipients –who must engage with critical controllers constantly. This project expects to generate new knowledge in design by esta ....Design guidelines for safety-critical controllers in high-risk environments. This project aims to generate novel product design guidelines for developing safer controllers for use by potentially stressed individuals in high-risk situations. It will do this by generating specific insights and verifying generalisable solutions from the context of total artificial heart recipients –who must engage with critical controllers constantly. This project expects to generate new knowledge in design by establishing a new research topic around an under-examined user cohort. Expected outcomes of this project include interaction design theory developments and improved controller design techniques. This should provide significant benefits and competitive advantages by lowering stress and improving safety across a range of contexts.Read moreRead less
Neuroimage as biomechanical model: new real-time computational biomechanics of the brain. This project is to extend to medicine the success computational mechanics has enjoyed in traditional engineering. The project will create enabling modelling and computing technologies for Computer-Integrated Surgery Systems that could help to improve clinical outcomes and the efficiency of health care delivery.