Can an automated vehicle with artificial intelligence system adapt to Australian road conditions? Can an AI reduce road mishaps? That is what researchers from Brisbane’s Queensland University of Technology (QUT) aim to figure out in their upcoming research project.
In order to determine this, QUT will go on a 1,200-km road trip using an electric car equipped with high-tech Artificial Intelligence sensors and computers. The trial is expected to last for three months as the road trip will be across Queensland.
In line with the university’s partnership with the State Government, the study will look at how the automated vehicle’s AI system adapts to Australian road conditions in four main areas:
- lane markings
- traffic lights
- street signage
- overcoming the limitations of GPS systems in built-up areas and tunnels for vehicle positioning.
The government is pushing the Cooperative and Automated Vehicle Initiative (CAVI) to help prepare for the arrival of new vehicle technologies with safety, mobility and environmental benefits on Queensland roads.
CAVI, which will be delivered by the Department of Transport and Main Roads, lays the technical foundations for the next generation of smart transport infrastructure. Transport and Main Roads Minister Mark Bailey said the road trip, which will be made in a zero- emissions Renault ZOE, will map Queensland roads for the cars of the future.
Challenges of Self-Driving Cars
The biggest disputes for the current generation of automated vehicles are driving and people, points out Professor Michael Milford from the university’s Australian Centre for Robotic Vision.
Professor Milford said early testing of the system has already revealed how a paint spill on the road could confuse a self-driving AI system into wrongly identifying it as a lane marking.
In fact, QUT’s initial experiments showed how automated cars have difficulties on rural roads which can lack lane markings.
“People will also cross the imaginary line to go around obstacles. It’s quite difficult for an automated vehicle to do this,” he said in a media statement.
“The primary goal of this work is to consider how current advances in robotic vision and machine learning – the backbone of AI – enable the research car platform to see and make sense of everyday road signage and markings that we, as humans, take for granted,” he added.
Regardless of the challenges, the Department of Transport and Main Roads still has high hopes for AI-equipped vehicles. Minister Bailey believes artificial intelligence technology and smart road infrastructure have potential to transform the way people travel in Queensland and reduce road trauma.
“This is world-leading transport technology research and it’s happening right here in Queensland,” Mr Bailey said.
For more information about the project, visit the Transport and Main Roads department website.