Machine learning is popping up in a range of different sports, helping to predict everything from athletes' injuries to peak performance levels. Enter the Tour de France -- the world's biggest cycling event, consisting of 198 riders across 22 teams who must traverse a total distance of 3,540 kilometres -- which is utilizing AI for the first time ever during this year's event.
The 104th edition of the race will see the Tour carry out a pilot machine learning program that will aim to predict the likelihood of various race scenarios. For example, the data could help researchers glean whether the peloton (the main pack of riders) will catch the breakaway riders at certain stages of the race.
Using GPS transponders, installed under the saddles of each bike, a whopping 3 billion data points will be collected throughout the 21 stages of the Tour. These insights will be combined with external data (such as the course gradient and weather conditions) to bring viewers a range of breaking stats, including live speed and the location of individual riders, distance between riders, and composition of groups within the race.
The hub for this information will be a cloud-based data centre, which will relay stats to broadcasters, allowing them to tell you even more about your favorite teams or riders. The Tour de France kicks off this Saturday July 1 in Düsseldorf, Germany, and concludes just over three weeks later at the Paris Champs-Élysées on Sunday July 23.