Latest News

How autonomous systems react in a changing environment

environment

Researchers from the Queensland University of Technology (QUT) Centre for Robotics have been awarded a $614,000 grant to investigate how robotics and autonomous vehicles can better respond to their environment. 

The University of Adelaide’s Centre for Advanced Defence Research in Robotics and Autonomous Systems (CADR RAS) awarded the grant, involving QUT’s Professor Michael Milford and Helen Carson, and Professor Iman Shames of the Australian National University. 

The project will analyse how autonomous systems navigate in challenging environmental conditions and detect and deal with deliberate attempts by “adversaries” to interfere with their navigation systems. 

“Positioning systems for robots performing critically important operations are useless unless they can adapt to changing environmental conditions, and respond to deliberate interference by adversaries,” Milford said, who is joint director at the Centre for Robotics. 

“This project aims to give robot and autonomous navigation systems the ability to detect and react to interference from adversaries, as well to adverse conditions in the natural environment. It’s an exciting team bringing together a roboticist, a control theorist and an aerospace and autonomous vehicle industry veteran, and we can’t wait to get started.” 

As a PhD researcher, Carson spent 20 years working in Defence on major aerospace projects before joining a Silicon Valley start-up, building autonomous robo-taxis. 

“Guaranteeing high integrity navigation performance under all conditions is key to deploying fully autonomous robotics — it’s a problem that has to be solved before we can truly remove human supervision,”Carson said. 

“I’m really eager to develop solutions that will unlock our ability to deploy autonomous systems in the real world.” 

Shames’ area of expertise include the mathematical aspects of control theory. 

“We aim to untangle the hacks that have dominated the field, so we can formally understand the conditions under which modern positioning algorithms work,” Shames said. 

Send this to a friend