Autonomous Precision Access: Agent Modelling for Cooperative Target Surveillance

Location: Melbourne, VIC

Duration: 5 months

Please note: Due to the sensitivity and security of this project, students must have Australian Citizenship to apply. Any applicants not meeting this requirement will automatically be deemed ineligible for this project.

Project Background

Unmanned Aerial Systems (UAS) agents undertaking Intelligence Surveillance and Reconnaissance missions such as Cooperative Target Observation (CTO) must account for the myriad challenges found in a hostile urban Defence context. These challenges include: visual obscuration, intermittent communication, environmental dynamics, partial observability, radio-frequency and electronic-warfare disruption and denial. To enable autonomous systems to cope with this complexity, DST Group is investigating approaches to embedding agents with a Theory of Mind (ToM) capability to better enable them to understand their peers and their operating environment. DST anticipates that an embodied ToM will allow agents to make meaningful inferences about the current and likely future actions of their peers, and to make meaningful inferences about unknowns based on observing peer behaviour. From this, agents will be able to better act in accordance with the context.

Research to be Conducted

The CTO scenario involves sensor positioning/emplacement strategies with an aim to maintain an efficient and continual coverage of selected targets, within the constraints of limited communication and partial observability. Agent Modelling can prove useful in such scenarios where a complex and communication-constrained environment results in incomplete and/or out-of-date information. The CTO problem encompasses the following set of sub-missions: (i) Sensor Pre-emplacement – agents deployed and travelling to their emplacement location; (ii) Target Assignment – assigning agents to the targets; and (iii) Sensor re-emplacement – sensor positioning strategy while accounting for the environmental dynamics. The research to be conducted includes the following:
• Investigating how Theory of Mind (more specifically Bayesian Theory of Mind) techniques can be used to reason about the beliefs of other agents which then will be used to predict their actions; and implementing the reasoning framework for sample scenarios.
• Investigating how Theory of Mind and Abductive Reasoning can be used to reason about the behaviour of other agents or situations which then will be used to plan its own behaviour or influence other agents’ plans; and implementing the reasoning framework for sample scenarios.

Skills Required

If you’re a PhD student and meet some or all the below we want to hear from you. We strongly encourage women, indigenous and disadvantaged candidates to apply:

  • Probability Theory (Conditional Probability, Bayes’)
  • Software Development in Java, C++
  • Agent Reasoning Techniques – Previous experience in application of reasoning techniques especially Abductive Reasoning Paradigm to derive most likely explanation from the set of uncertain/incomplete observations. This experience although not mandatory will be highly regarded.

Expected Outcomes

The anticipated outcomes are:
• The candidate contributes to the research and implementation of the Theory-of-Mind and Reasoning approaches in a simple scenario set as agreed.
• The candidate contributes to the development of software-based simulation environment for implementing simulation scenarios and testing the approach.
• Publication/Report describing the implementation details of the algorithm and performance of the algorithm subject to variations in the simulation parameters

Additional Details

The intern will receive $3,000 per month of the internship, usually in the form of stipend payments.

It is expected that the intern will primarily undertake this research project during regular business hours, spending at least 80% of their time on-site with the industry partner.  The intern will be expected to maintain contact with their academic mentor throughout the internship either through face-to-face or phone meetings as appropriate.

The intern and their academic mentor will have the opportunity to negotiate the project’s scope, milestones and timeline during the project planning stage.

Applications Close

27 November 2019

Reference

APR – 1262