Developing Adversarial Reinforcement Learning Systems for Swarm Defense

About this project

Improve defence systems against drone attacks

If you are seeking to build a career in the Defence sector and are passionate about AI research with real world applications, the University of South Australia – Australia’s University of Enterprise – is offering a hands-on project-based PhD within the Australian Research Centre for Interactive and Virtual Environments (IVE), in partnership with Acacia Systems.  

The goal of the project is to implement an adversarial reinforcement learning (ARL) model to optimise the performance of a defensive suite of sensor and effectors implemented to counter a drone threat. A simulation model will be developed to reflect the defensive reinforcement learning (RL) agents i.e. the suite of sensors and effector defending against a drone attack and the offensive RL agents i.e. the swarm of attacking drones.  

A 3D visualisation tool will also be developed to clearly show the real time output of the simulation model. The ARL model of the offensive and defensive agents will compete against each with the goal of developing the optimal defensive strategy. The system will recommend which effector should engage which drone and in which order. This ARL approach will be then compared to other current threat evaluation systems and iteratively improved until the outcome is significantly better. 

This PhD position is made possible by the CSIRO Next Generation Graduate Program, and as part of the program students will receive addition benefits, including an industry internship, special course on emerging technology, and additional travel and research funding. For full details, see the CSIRO Next Generation Graduate Program.

What you’ll do

In this project-based research degree, you will try to support Acacia Systems’ work in developing AI systems for defending against a swarm of drones. Acacia Systems have developed several command-and-control systems for drone defence and would like to include AI systems as part of those.  

Upon completion of the project, you will have developed new AI algorithms and visualisation techniques, published articles and developed prototypes and demonstrations of working systems. 

You will have the opportunity to undertake a 6-month industry internship with Acacia Systems to get hands-on practical industry experience. 

Where you’ll be based

This is a joint project between the Australian Research Centre for Interactive and Virtual Environments (IVE), and the Industrial AI Research Centre (IAI) at the University of South Australia. IVE is a world leader in Augmented Reality (AR) and Virtual Reality (VR), conducting research in many areas of AR and VR including wearable computing, interface design, empathic computing, 3D visualisation, perception, and telepresence.  

IAI brings together experts from across computer science to develop solutions for autonomous and augmented intelligence systems. Staff and students from IAI are a part of the global revolution in artificial intelligence, machine learning, Industry 4.0 and Internet-of-Things (IoT) technologies.

The student will be jointly supervised by faculty from both research centres and be based in the IVE research centre building next to IAI. Financial Support

This project is funded for reasonable research expenses. Additionally, a living allowance scholarship of $41,650 per annum is available to Australian or New Zealand citizens and Australian permanent residents. Australian Aboriginal and/or Torres Strait Islander applicants will be eligible to receive an increased stipend rate of $50,291 per annum (2024 rates). A fee-offset or waiver for the standard term of the program is also included. For full terms and benefits of the scholarship please refer to our scholarship information.

Eligibility and Selection

This project is open to applications from Australian or New Zealand citizens and Australian permanent residents. International applicants are not invited to apply at this time.

Applicants must meet the eligibility criteria for entrance into a PhD. Additionally, applicants must meet the project selection criteria: 

• Background or interest in Artificial Intelligence, Machine Learning, Reinforcement Learning and/or Visualisation

All applications that meet the eligibility and selection criteria will be considered for this project. A merit selection process will be used to determine the successful candidate.

The successful applicant is expected to study full-time and to be based at our Mawson Lakes campus in the north of Adelaide.

Essential Dates

Applicants are expected to start in a timely fashion upon receipt of an offer. Extended deferral periods are not available. Applications close on Tuesday 5 November 2024.