The team of the Vienna University of Technology took first place in the race. Sophisticated computer software tipped the scales. The vehicles have to complete their laps without human intervention during the race. The artificial intelligence developed by the individual teams before the race have to correctly recognise the route, avoid collisions and drive fast lap times. Cameras and sensors provide the necessary data, which is then interpreted by the autonomous software in real time and converted into control commands for the model car. After a total of three competitive rounds, the "Scuderia Segfault" of the Vienna University of Technology prevailed as the overall winner.
"The trajectory and the velocity profile are optimised offline to enable the laps to be driven as quickly as possible," said Luigi Berducci of the Vienna University of Technology team. "With the help of machine learning, a series of promising parameters are determined which are subsequently optimised with computer simulations and ultimately validated on the real vehicle," he added.
F1TENTH Autonomous Racing is a competition organised by an international community of researchers, engineers and enthusiasts of autonomous systems. The race took place within the context of the 2021 IEEE/RSJ International Conference on Intelligent Robots and Systems in Prague.
Go on reading and learn more about mobility research in Austria!