World-1st roadside AI tech that prevents animal-vehicle collisions tested

A team of researchers from the University of Sydney, Queensland University of Technology (QUT), and the Department of Transport and Main Roads Queensland has developed and successfully tested the world’s first roadside AI technology designed to prevent animal-vehicle collisions. Known as the Large Animal Activated Roadside Monitoring and Alert (LAARMA) system, it uses a combination of RGB cameras, thermal imaging, LiDAR sensors, and self-teaching artificial intelligence to detect animals near roads in real-time and alert drivers through flashing Variable Message Signs (VMS). During a five-month trial in Far North Queensland, an area with frequent cassowary collisions, LAARMA achieved 97% detection accuracy, recorded over 287 animal sightings, and helped reduce vehicle speeds by up to 6.3 km/h. Unlike traditional detection systems, LAARMA’s AI continuously improves its accuracy by learning from each sighting without human reprogramming, increasing detection rates from an initial 4.2% to 78.5%