Unifying Foundations for Intelligent Agents. This project aims to drive forward the development of rigorous foundations for intelligent agents. The agent framework, the expected utility principle, sequential decision theory, and the information-theoretic foundations of inductive reasoning and machine learning have already brought significant order into the previously heterogeneous scattered field of artificial intelligence. This project aims to investigate an information-theoretic approach towar ....Unifying Foundations for Intelligent Agents. This project aims to drive forward the development of rigorous foundations for intelligent agents. The agent framework, the expected utility principle, sequential decision theory, and the information-theoretic foundations of inductive reasoning and machine learning have already brought significant order into the previously heterogeneous scattered field of artificial intelligence. This project aims to investigate an information-theoretic approach towards a unifying foundation for intelligent agents, which has recently spawned impressive applications. The theory is expected to provide a gold standard and valuable guidance for researchers working on smart software.Read moreRead less
Feature reinforcement learning. Agent applications include speech recognition systems, vision systems, search engines, auto-pilots, spam filters, and robots. The research outputs from this project will enable agents to adapt to their environment and automatically, during deployment, acquire much of the knowledge that is currently required to be built in by agent designers.
On-line planning for constrained autonomous agents in an uncertain world. This project aims to develop admissible heuristics for constrained stochastic planning problems and integrate them into state-of-the-art on-line algorithms. Artificial Intelligence (AI) systems, such as self-driving cars or energy management systems, make intelligent decisions to act near-optimally in uncertain environments, leading to savings, for example in energy use. But we also want assurances that AI systems will obe ....On-line planning for constrained autonomous agents in an uncertain world. This project aims to develop admissible heuristics for constrained stochastic planning problems and integrate them into state-of-the-art on-line algorithms. Artificial Intelligence (AI) systems, such as self-driving cars or energy management systems, make intelligent decisions to act near-optimally in uncertain environments, leading to savings, for example in energy use. But we also want assurances that AI systems will obey safety constraints. Solving constrained stochastic planning problems is key to building AI that is both robust and safe. This project will explore ways to account for uncertainty and constraints in heuristic estimation.Read moreRead less
Lifelong robotic navigation using visual perception. Service robots are becoming a major part of our working and personal environments, in much the same way as personal computers already have. This project will develop new methods of practical and useful robot navigation that will enable Australia's industries and services to remain internationally competitive.
Navigating brains: the neurobiology of spatial cognition. Navigation is one of the most crucial and most challenging problems animals face. Behavioural analyses have shown that animals make use of a number of different mechanisms to navigate, but very little is known of how different forms of spatial information are processed and integrated by the brain. The project aims to tackle this by placing tethered ants in a virtual-reality simulation of their real environment allowing precise control of ....Navigating brains: the neurobiology of spatial cognition. Navigation is one of the most crucial and most challenging problems animals face. Behavioural analyses have shown that animals make use of a number of different mechanisms to navigate, but very little is known of how different forms of spatial information are processed and integrated by the brain. The project aims to tackle this by placing tethered ants in a virtual-reality simulation of their real environment allowing precise control of visual navigational cues, as well as the opportunity to study the brains of the tethered ants as they solve the real-world challenge of finding home. This may reveal how simple brains efficiently solve navigational tasks, which may inform both cognitive biology and bio-inspired computation.Read moreRead less
AI Planning: The Next Generation. This is a project in Artificial Intelligence. It aims at extending and integrating automated planning (and other forms of reasoning) with learning to produce a new generation of planning systems that are robust, safe, scalable, and trusted. These are some of the most significant issues to address to accelerate the adoption of planning systems in industry. Expected outcomes include a pipeline to learn rich symbolic planning models from narrated demonstration vide ....AI Planning: The Next Generation. This is a project in Artificial Intelligence. It aims at extending and integrating automated planning (and other forms of reasoning) with learning to produce a new generation of planning systems that are robust, safe, scalable, and trusted. These are some of the most significant issues to address to accelerate the adoption of planning systems in industry. Expected outcomes include a pipeline to learn rich symbolic planning models from narrated demonstration videos, new ways to represent, learn, and search for generalised policies that are scalable and robust, and approaches to verify and explain generalised policies. The new systems should benefit the aerospace industry by assisting humans in assembling and delivering aerospace products.Read moreRead less
Acquiring and Using Views for Homing. The aim of the project is to investigate how insects acquire and use scene memories for homing, a crucially important task for most animals. In bees and wasps, these memories are acquired during learning flights when leaving the nest. This fast, active learning process underpins much of the insects' navigational competence, but it remains unknown how it is controlled and how in detail memories guide returns to the nest. It is intended to use the latest camer ....Acquiring and Using Views for Homing. The aim of the project is to investigate how insects acquire and use scene memories for homing, a crucially important task for most animals. In bees and wasps, these memories are acquired during learning flights when leaving the nest. This fast, active learning process underpins much of the insects' navigational competence, but it remains unknown how it is controlled and how in detail memories guide returns to the nest. It is intended to use the latest camera-based reconstruction tools for the first time: to quantify the navigational information content of habitats including the visual information available to learning and homing insects, and to dynamically modify the insects’ natural visual environment in order to critically test hypotheses about acquisition and use of views for homing.Read moreRead less
Exploiting the Symmetry of Spatial Awareness for 21st Century Automation. This project aims to enable autonomous robotic systems to operate more robustly and more reliably in the complex, cluttered and dynamic environments found in real-world applications. Applying the latest understanding of symmetry in non-linear systems and control provides tools that can be used to develop new design methodologies for spatial awareness algorithms. The outcomes of this project should increase Australia's ca ....Exploiting the Symmetry of Spatial Awareness for 21st Century Automation. This project aims to enable autonomous robotic systems to operate more robustly and more reliably in the complex, cluttered and dynamic environments found in real-world applications. Applying the latest understanding of symmetry in non-linear systems and control provides tools that can be used to develop new design methodologies for spatial awareness algorithms. The outcomes of this project should increase Australia's capacity in high-tech systems and deliver world best open source code for spatial awareness problems to enable the next generation of automation in Australia.Read moreRead less
Structure-without-motion: large-scale 3D reconstruction from distributed and unorganised images. Vision-based 3D reconstruction is a frontier technology for a wide range of applications. This project will lead to novel 3D reconstruction methods and systems that are more efficient, more cost-effective and more accessible to ordinary user. The outcomes will directly contribute to National Research Priority Goal of smart information use.
Integrated high-performance control of aerial robots in dynamic environments. The outcomes of this project will enable novice pilots to safely operate aerial robots in dangerous and dynamic environments through novel intuitive user interfaces and advanced control algorithms. The project will contribute strongly to Australia's presence in the emerging world market of unmanned aerial vehicles.