ITS Australia: vehicle data integration can 'transform safety'

New research offers an “Australian-first demonstration” of how multi-source data integration and emerging technologies can improve traffic signal efficiency using AI and connected vehicle data.

A report, based on research from ITS Australia, is called Integrated Connected Data for Safer, More Efficient Transport Management. The project explores how connected vehicle data, bicycle telemetry, artificial intelligence and real-time analytics can “transform the safety, efficiency and sustainability of our transport systems”.

The research was funded through iMove CRC and supported by the Australian government's CRC programme. It was delivered in partnership with the University of Melbourne, Transport for New South Wales, Department of Transport and Planning Victoria, Department of Transport and Main Roads Queensland, Main Roads Western Australia and the Transport Accident Commission.

The research proactively identifies safety risks using gyroscopic and braking behaviour data. Accurate mapping can be done of congestion and emissions in real time using vehicle trajectory analysis. Meanwhile, cyclist safety and infrastructure performance can be enhanced through micromobility telemetry.

The research noted that, unlike simulations, real vehicle trajectories captured nuanced traffic-emission relationships. This highlighted the importance of managing congestion and intersection approaches.

Deep reinforcement learning, computer vision and ensemble AI models demonstrated strong potential to predict road conflicts and optimise signal timing, outperforming conventional systems.

Incorporating cyclist movement and braking data revealed hidden safety risks and flow inefficiencies. This provided actionable insights for intersection and infrastructure design.