Neuro-Autonomy

Neuro-inspired perception, navigation and spatial awareness for autonomous robots

An Australia-US Multidisciplinary University Research Initiative (AUSMURI)
Funded by the Australian Government

Project overview

State-of-the-art autonomous vehicles (AVs) are trained for specific, well-structured environments and, in general, would fail to operate in unstructured or novel settings. This project aims at developing next-generation AVs, capable of learning and on-the-fly adaptation to environmental novelty. These systems need to be orders of magnitude more energy efficient than current systems and able to pursue complex goals in highly dynamic and even adversarial environments.

Biological organisms exhibit the capabilities envisioned for next-generation AVs. From insects to birds, rodents and humans, one can observe the fusing of multiple sensor modalities, spatial awareness, and spatial memory, all functioning together as a suite of perceptual modalities that enable navigation in unstructured and complex environments. With this motivation, the project will leverage deep neurophysiological insights from the living world to develop new neuroscience-inspired methods capable of achieving advanced, next-generation perception and navigation for AVs.

The project will advance a science of autonomy, which is critical in enhancing capabilities to execute missions using ground, sea, and aerial AVs.

This is a multidisciplinary collaboration, spanning automatic control, biology, machine learning, neuro-engineering, robotics, and signal processing. It involves investigators at the University of Melbourne, Macquarie University, the University of New South Wales and Queensland University of Technology, working closely with an allied MURI project at Boston University and MIT.

Research activities are organised around four tightly coupled thrusts:

  • Thrust I activities consist of observing and recording animal spatial awareness and navigation capabilities, correlating these with neuronal activity, and developing explanatory/predictive computational models. Animal and computer-based experiments will be conducted.
  • Thrust II will develop biologically-inspired methods for perception to enable goal-directed navigation. It will incorporate semantic representations, control actions to enhance perception (eg, vision angle), metrics of information gain, detecting changes in the environment, and visual place perception. A key output of this thrust will be a suite of perceptive features to be used for developing autonomous navigation capabilities.
  • Thrust III will develop next-generation AV capabilities via unified knowledge representation, control, and learning strategies. The research will use features from Thrust II, determine how to represent the environment and past experiences, develop methods to extract navigation control policies from animal and human observations, and optimize these policies to achieve autonomous goal-directed navigation in new settings.
  • Thrust IV will leverage a plethora of experimental testbeds at participating institutions to design experiments that match animal/human experiments under Thrust I. The purpose is to test the effectiveness and efficiency of methods developed under the project in laboratory and real-world settings. Lessons from this Thrust will provide feedback to methodology development and motivate additional experiments.

News


The University of Melbourne investigators

For more information, contact the following investigators:

Girish Nair

Girish Nair is a Professor in the Department of Electrical and Electronic Engineering. He is a Fellow of the IEEE (2018) and has received several prizes, including the 2014 George S. Axelby Outstanding Paper Award from the IEEE Control Systems Society, a 2006 SIAM Outstanding Paper Prize, the Best Theory Paper Prize at the UKACC International Conference on Control, Cambridge University, 2000, and the L.R. East Medal for Best Performance in Final Year Engineering, 1994. From 2015–2019 he was an Australia Research Council Future Fellow. His research interests focus on the interplay between feedback control and information theory, in both stochastic and nonstochastic settings.

Girish is the Australian Principal Investigator of the AUSMURI team and can be contacted at gnair@unimelb.edu.au

Personal website


Anthony Burkitt

Professor Anthony Burkitt holds the Chair in Bio-Signals and Bio-Systems in the Department of Biomedical Engineering at the University of Melbourne. Prof Burkitt's main research interests are in understanding how the brain processes and learns information. His research in computational neuroscience has contributed to understanding the behaviour and function of neural information processing in the brain, encompassing both neural coding and spike-timing synaptic plasticity. Professor Burkitt was the Director of Bionic Vision Australia (2010-2016), a Special Research Initiative in Bionic Vision Science and Technology of the Australian Research Council (ARC). He successfully led the project through all of its phases: Project conception, securing $50 million in ARC funding, the research and development of a prototype bionic eye (suprachoroidal retinal implant), the successful implantation in three patients, and the establishment of the company Bionic Vision Technologies (BVT) with US$18 million of venture capital for the ongoing clinical and commercial development of the technology. Professor Burkitt’s research encompasses also the areas of neuro-engineering, medical bionics, cochlear-implant speech processing and bio-signal processing for epilepsy. His research has been instrumental in the development of visual stimulation paradigms for retinal implants, new cochlear implant speech processing strategies, methods for detecting and predicting seizures, and the use of electrical stimulation for seizure abatement in epilepsy.

Anthony can be contacted at aburkitt@unimelb.edu.au.


William Moran

Bill Moran (M’95) currently serves, since 2017, as Professor of Defence Technology in the University of Melbourne. From 2014 to 2017, he was Director of the Signal Processing and Sensor Control Group in the School of Engineering at RMIT University, from 2001 to 2014, a Professor in the Department of Electrical Engineering, University of Melbourne, Director of Defence Science Institute in University of Melbourne (2011-14), Professor of Mathematics (1976–1991), Head of the Department of Pure Mathematics (1977–79, 1984–86), Dean of Mathematical and Computer Sciences (1981, 1982, 1989) at the University of Adelaide, and Head of the Mathematics Discipline at the Flinders University of South Australia (1991–95). He was Head of the Medical Signal Processing Program (1995–99) in the Cooperative Research Centre for Sensor Signal and information Processing. He was a member of the Australian Research Council College of Experts from 2007 to 2009. He was elected to the Fellowship of the Australian Academy of Science in 1984. He holds a PhD in Pure Mathematics from the University of Sheffield, UK (1968), and a First Class Honours B.Sc. in Mathematics from the University of Birmingham (1965). He has been a Principal Investigator on numerous research grants and contracts, in areas spanning pure mathematics to radar development, from both Australian and US Research Funding Agencies, including DARPA, AFOSR, AFRL, Australian Research Council (ARC), Australian Department of Education, Science and Training, and Defence Science and Technology, Australia. His main areas of research interest are in signal processing both theoretically and in applications to radar, waveform design and radar theory, sensor networks, and sensor management. He also works in various areas of mathematics including harmonic analysis, representation theory, and number theory.

Bill can be contacted at wmoran@unimelb.edu.au.


Iman Shames

Iman Shames is an Associate Professor at the University of Melbourne. Previously, he had been a Senior Lecturer and McKenzie Research Fellow at the department of Electrical and Electronic Engineering, the University of Melbourne from 2013.

He was an ACCESS Postdoctoral Researcher at the ACCESS Linnaeus Centre, the KTH Royal Institute of Technology, Stockholm, Sweden from 2011 to 2012. He received his PhD degree in engineering and computer science from the Australian National University, Canberra, Australia in 2011 and his BS degree in Electrical Engineering from Shiraz University in 2006. His current research interests include, but are not limited to, optimisation theory and its application in control and estimation, mathematical systems theory, and security and privacy in cyber-physical systems.

Iman can be contacted at iman.shames@unimelb.edu.au.

Personal website


Other Australian investigators

Ken Cheng

Macquarie University

Ken Cheng has studied animal behaviour for some 40 years. A large part of the research has been and continues to be on spatial behaviour. Over his career, he has studied humans, monkeys, rats, a number of species of birds, honeybees, desert ants, and bull ants. Starting this century, Cheng’s team has been studying navigation and learning in an Australian desert ant located in Central Australia. Known as the red honey ant (the left two photos in the banner above), Melophorus bagoti shares many characteristics with the much-studied North African desert ants of the genus Cataglyphis. It is long-legged, active in the heat of the day, moves fast, and deploys a suite of navigational strategies. Cheng is the author of three books, a reference book for a general audience on animal cognition, a textbook on the biological basis of behaviour, and a small, short guide on scientific writing, aimed at postgraduate students or early-career scientists.

Ken can be contacted at ken.cheng@mq.edu.au.


Michael Milford

Queensland University of Technology

Professor Milford conducts interdisciplinary research at the boundary between robotics, neuroscience and computer vision and is a multi-award-winning educational entrepreneur. His research models the neural mechanisms in the brain underlying tasks like navigation and perception to develop new technologies in challenging application domains such as all-weather, anytime positioning for autonomous vehicles. He is also one of Australia’s most in demand experts in technologies including self-driving cars, robotics and artificial intelligence, and is a passionate science communicator. He currently holds the position of Professor at the Queensland University of Technology, as well as Deputy Director of the QUT Centre for Robotics, Microsoft Research Faculty Fellow and Chief Investigator at the Australian Centre for Robotic Vision.

Michael can be contacted at michael.milford@qut.edu.au.


Andrey Savkin

UNSW

Andrey V. Savkin was born in 1965 in Norilsk, USSR. He received his MS degree in 1987 and PhD degree in 1991, both from the Leningrad State University. In 1987–1992, he worked in the All-Union Television Research Institute as an engineer and as a Senior Research Scientist. In 1992–1994, he was a Postdoctoral Research Fellow with the Department of Electrical Engineering, Australian Defence Force Academy. In 1992–1994, Savkin was a Research Fellow with the Department of Electrical Engineering, the University of Melbourne. In 1996–2000, he was a Senior Lecturer and then an Associate Professor with the Department of Electrical and Electronic Engineering, the University of Western Australia. Andrey Savkin is a Professor with the School of Electrical Engineering and Telecommunications, the University of New South Wales, Sydney, Australia since 2000. He has authored and co-authored 7 research monograph (published by Springer, Birkhauser, IEEE Press – Wiley and Elsevier) and about 240 journal papers. Prof Savkin served as an Associate Editor/Editor for numerous international journals conferences in the field. His current research interests include control engineering, robotics, power systems, sensor networks and biomedical engineering.

Andrey can be contacted at a.savkin@unsw.edu.au.


Research Fellows

Tobias Fischer

Queensland University of Technology

Dr Tobias Fischer conducts interdisciplinary research at the intersection of computer vision, cognitive robotics and computational cognition. His main goal is to develop high-performing, bio-inspired computer vision algorithms that can be simultaneously used to examine the perceptional capabilities of animals/humans and robots. Before joining QUT as a Research Fellow in January 2020, Dr Fischer was a postdoctoral researcher in the Personal Robotics Lab at Imperial College London. He received a PhD from Imperial College in January 2019. Dr Fischer’s thesis has been awarded the UK Best Thesis in Robotics Award 2018 and the Eryl Cadwaladr Davies Prize for the best thesis in Imperial’s EEE Department in 2018. He previously received an M.Sc. degree (distinction) in Artificial Intelligence from The University of Edinburgh, in 2014, and a BSc degree in Computer Engineering from Ilmenau University of Technology, Germany, in 2013. His papers have been awarded two best poster awards, one at the Samsung AI Forum 2018 and one at the Gaze Workshop at ICCV2019.

Tobias can be contacted at tobias.fischer@qut.edu.au.


Emilien Flayac

The University of Melbourne

Emilien Flayac has recently started as a Research Fellow at the University of Melbourne on the AUSMURI project Research Fellow at the Department of Electrical and Electronic Engineering, the University of Melbourne. He received his BSc degree in Optimisation and Control from Paris-Saclay University in 2016, and his PhD degree in Applied Mathematics from Paris-Saclay University in France in 2019. His current research interests include nonlinear observers, particle filtering, dual control and stochastic model predictive control.

Emilien can be contacted at emilien.flayac@unimelb.edu.au.


Muzhid Islam

Macquarie University

Muzhid Islam is a Research Associate in the department of biological sciences at Macquarie University. He has recently completed his PhD program in Behavioural Neuroscience at Macquarie University, Sydney with Professor Ken Cheng. He is interested in how behaviour is generated by the interaction of the brain, body, and environment. His PhD research is about learning, decision-making, and visual navigation in an insect which is conducive to developing computational models of the underlying mechanisms of learning, an essential goal in developing autonomously navigating systems in the field of Bio-robotics. He is currently investigating the neuronal, sensory and visual basis of self-control during navigation in Australian bull ants and how they produce visual scans in their natural environment and what this self-control and scans contribute to the individual learning and decision-making.

Muzhid can be contacted at muzahid.islam@mq.edu.au


Yanbo Lian

The University of Melbourne

Yanbo Lian is a Post-doc Research Fellow in Neuro-engineering in the Department of Biomedical Engineering at the University of Melbourne. He received his PhD degree in computational neuroscience from the University of Melbourne in 2020. His PhD thesis is about building learning models to understand biological vision processing using concepts from machine learning. His main research interests are in understanding how the brain processes information and how the brain functions arise from learning.

Yanbo can be contacted at yanbo.lian@unimelb.edu.au.


Tim Molloy

The University of Melbourne

Tim received his BEng (Aerospace Avionics) and PhD degrees from the Queensland University of Technology (QUT) in 2010 and 2015, respectively. From 2016 to 2019 he was a Research Fellow and then Advance Queensland Research Fellow at QUT investigating inverse dynamic game theory, quickest detection, and robotic vision for unmanned aircraft collision avoidance. Since 2020 he has been a Research Fellow at the University of Melbourne on the Neuro-Autonomy AUSMURI project. Tim is the recipient of a QUT University Medal, a QUT Outstanding Doctoral Thesis Award, and an Advance Queensland Early Career Research Fellowship supported by Boeing Research & Technology Australia. His research interests include information theory, control, and signal processing for robots and autonomous systems.

Tim can be contacted at tim.molloy@unimelb.edu.au.


Trevor Murray

Macquarie University

Dr Murray is a neuro-ecologist that specialises in integrating computational; audio, visual, and locomotor recording; and experimental techniques to study animal behaviour, its neuronal basis, and its applications for other fields. His recent work on visual navigation and learning involves multi-disciplinary projects, which allow him to apply his expertise with sensory, neural and behavioural ecology to problems in the domains of artificial and robotic autonomous navigation. As part of his postdoctoral work at Australian National University, he collaboratively developed an insect-catered virtual reality system for studying the navigational capabilities and processes of Australian bull ants.

Trevor can be contacted at trevor.murray@mq.edu.au.


Mohsen Eskandari

University of New South Wales

Mohsen Eskandari was born in Saveh, Iran. He received the B.Sc. and M.Sc. degrees from the Islamic Azad University, Saveh Branch, Saveh, Iran, in 2004 and 2013, respectively, and the Ph.D. degree from the University of Technology Sydney, Sydney, Australia, in 2021, all in Electrical Engineering. Since March 2020, he has been a Postdoctoral Research Associate with the University of New South Wales, Sydney, Australia. He joined the AUSMURI project as a Research Fellow early 2021.

Mohsen has more than ten years of experience in different parts of the electrical industry. He has proven skills in handling power systems projects as well as a strong background in the field of automation and control. His research interests include power systems, power electronics, micro/smart grids, control theory and optimisation, and recently AI and Navigation.

Mohsen can be contacted at m.eskandari@unsw.edu.au


PhD students

Marvin Chancán

Queensland University of Technology

Marvin Chancán is a PhD candidate in the School of Electrical Engineering and Robotics at Queensland University of Technology (QUT). He is leading research at the intersection of machine learning, neuroscience, reinforcement learning, and computer vision, with the aim of developing learning algorithms and techniques that can endow robotic systems with the required skills for executing diverse tasks in the real world. In particular, Marvin is interested in how learning capabilities, found both in biological brains and artificial systems, can be used to acquire complex behaviours such as visual localisation and navigation for enabling greater autonomy and intelligence in robots and autonomous vehicles. He is the recipient of a QUT HDR High Achiever Award that recognises sustained levels of high achievement as evidenced by his outstanding research outputs produced during the first two years of his PhD. Marvin’s first paper of the PhD was published in the IEEE Robotics and Automation Letters journal and this was possible thanks to a multidisciplinary research collaboration that he initiated with his Harvard University collaborator in Systems Neuroscience, as well as a collaboration with Macquarie University. Before joining QUT as a PhD student, Marvin was a Lecturer with the Pontifical Catholic University of Peru and a Solutions Architect of high performance computing clusters in the IT industry from 2016 to 2018. Previously, he worked as a Mechatronics Engineer in the automation industry in Spain and Brazil from 2012 to 2015. He also received the Master’s degree in applied mechanics from the Pontifical Catholic University of Rio de Janeiro, Brazil, in 2012, and the B.Sc. degree in mechatronics engineering (graduated First in class) from Universidad Nacional de Ingeniería, Peru, in 2009. His research outcome can be found on Google Scholar.

Marvin can be contacted at marvin.chancan@hdr.qut.edu.au


Stephen Hausler

Queensland University of Technology

Stephen Hausler is a PhD student at Queensland University of Technology, Brisbane, Australia and is conducting research in autonomous navigation and computer vision. His research goal is to develop algorithms which enable an autonomous vehicle to localise and perceive its surroundings, only using a vision sensor and in potentially GPS denied environments. This research is inspired by the perceptual and navigation capabilities of animals and humans in conjunction with recently developed techniques in computer vision. He has several published research articles, including an article in IEEE Robotics and Automation Letters. He previously worked as an Electrical Engineer in a manufacturing environment and also graduated with a Bachelor of Electrical Engineering with First Class Honours in 2013.

Stephen can be contacted at stephen.hausler@hdr.qut.edu.au.


Vito Lionetti

Macquarie University

Vito Lionetti is a PhD Candidate in Biological Sciences at Macquarie University. In the recent years, he has been working on different projects on navigation behaviour in Australian bull ants and desert ants, in particular on the use, and learning of visual cues. As part of his PhD research topic, he is collaborating on a project focused on the use of virtual reality for navigational studies.

Vito can be contacted at lionetti.vitoantonio@gmail.com.


Maxwell Varley

The University of Melbourne

Max Varley is a PhD student in the Department of Electrical and Electronic Engineering at the University of Melbourne. He received his BSci(2017) and Mc-Eng(2019) in Electrical Engineering at the University of Melbourne. Max was part of a group that presented their year-long project Efficient Vision-Based Navigation at Low Bit-Rates at the University of Melbourne’s Endeavour Exhibition 2019 and received the award ‘Best Project in Electrical Engineering’.

He hopes to continue his research into efficient autonomous visual navigation while a part of the AUSMURI project.

Max can be contacted at varleym@student.unimelb.edu.au.


Yu Zhang

The University of Melbourne

Yu Zhang is a PhD student at the University of Melbourne. He received his Master's degree in biomedical engineering (2020) and BSc degree in Physics (2017) from the University of Melbourne. Yu is interested in neural coding and neuronal behaviour. His potential research direction is computational modeling of neuronal signaling in the sensory cortex.

Yu can be contacted at yuz9@student.unimelb.edu.au.


Jinghe Yang

The University of Melbourne

Jinghe Yang is a PhD student in the Department of Electrical and Electronic Engineering at the University of Melbourne. He received the M.Eng degree in Mechatronics (With Distinction) at the University of Melbourne in 2021 and the B.Eng degree in Mechanical Engineering at the Northeastern University, China, in 2017.

His research interests include Underwater SLAM and Model Predictive Control.

Jinghe can be contacted at jinghey@student.unimelb.edu.au


Somayeh Hussaini

Queensland University of Technology

Somayeh Hussaini is a PhD student at Queensland University of Technology, Brisbane, Australia. Her research interests are in the intersection areas of computer vision and neuroscience. The focus of her research is on developing re-evolving biological neural networks for spatially-informed intelligence.

She received her Bachelor of Engineering with First Class Honours in 2020 from Queensland University of Technology. She has been awarded the QUT Kindler Medal and the IEEE Best Final Year Thesis in All Fields of Electrical Engineering and Information Technology.

Somayeh can be contacted at somayeh.hussaini@hdr.qut.edu.au


Recruiting PhD students

We are looking for talented, motivated PhD students to join our project.

If interested, please email Girish Nair, gnair@unimelb.edu.au, or one of the other investigators directly with your CV, Bachelor’s and Master’s transcripts, and a brief research statement, with the subject heading “AUSMURI PhD”. Scholarships and fee remissions are available.