Australia’s Defence Science and Technology Group will host the Australian Defence Science, Technology and Research Summit (ADSTAR) in Adelaide from August 4 to six, 2026, bringing collectively leaders from authorities, academia, and trade to debate protection innovation.
One of the summit’s 4 key themes is “Integrated Ecosystem,” which focuses on how collaboration throughout sectors can speed up the event of protection capabilities. A current undertaking exploring AI-enabled training for air battle managers illustrates that idea in apply.
Developing AI brokers as sensible position gamers
Defence Science and Technology Group partnered with Macquarie University and Australian simulation firm PLEXSYS to develop AI brokers able to performing as sensible position gamers in Royal Australian Air Force training workout routines.
Air battle managers function in high-pressure command-and-control environments, integrating sensor knowledge, guiding plane, and coordinating missions. Their capability to make fast, correct choices is essential in fight situations.
The analysis workforce got down to decide whether or not AI may replicate the habits of professional human operators effectively sufficient to function efficient training companions. The AI brokers had been built-in into PLEXSYS’s Advanced Simulation Combat Operations Trainer, often called ASCOT 7.
Training the longer term pressure 💻
Defence innovation not often emerges from a single lab, workforce or organisation. It is created on the intersection of operational want, scientific curiosity and industrial functionality. When authorities, academia and trade work collectively, innovation… pic.twitter.com/0Vs9BXDpOy
— Defence Australia (@DefenceAust) May 18, 2026
Building the system
To develop the AI, researchers noticed skilled human operators working in simulated command-and-control settings. They studied how operators communicated, prioritized info, and made choices underneath stress.
The AI was then refined to reflect the habits, communication types, and decision-making patterns of high-performing groups. More not too long ago, the workforce included large-language fashions to assist the AI talk extra naturally with trainees.
Defense scientist Simon Hosking stated the undertaking developed considerably over time.
“We went from looking at whether we could train AI agents to replicate the behavior of expert operators to considering how to utilize the huge increase in the capability of large-language models to help the AI system communicate naturally with the trainees,” Hosking stated.
Feedback: Trainees couldn’t inform the distinction
During trials, contributors labored alongside each human position gamers and AI brokers. Feedback confirmed that trainees couldn’t reliably distinguish between the 2, indicating the system’s effectiveness in replicating human habits.
Hosking stated seeing the analysis transfer towards real-world utility was a milestone.
“To see our research transition from the lab to defense industry is a really proud thing to have achieved,” he stated.
Funding and improvement
The undertaking has been supported by way of a number of funding streams, together with Defence Science and Technology Group, Australia’s Next Generation Technologies Fund, and the Centre for Advanced Defence Research in Robotics and Autonomous Systems.
Most not too long ago, the work was prolonged underneath the Advanced Strategic Capabilities Accelerator by way of the Emerging and Disruptive Technologies – Decision Advantage program, with a contract valued at greater than $3 million. The extension is meant to additional develop the aptitude and align it with protection priorities.
Early-bird registration for the ADSTAR Summit closes May 29, 2026.