AI helped design bike that broke speed records

An aerodynamic bicycle has been used to set world records for top speed for cyclists. The bike was designed using software from a company called Neural Concepts, a spin-off of the Computer Vision Laboratory in the EPFL School of Computer and Communication Sciences. On September 13, a new world record for women's cycling speed was set.

Pilot Ilona Peltier biked down a 200-meter track at 126.5 kilometers per hour (78.6 mph) to set the world record. The day before she set the world record, her teammate Fabian Canal set a new men's university world record with a top speed of 136.7 km/h (84.9 mph). The speeds for the duo were recorded at the World Human Powered Speed Challenge in Nevada.

The bike the two rode is a pod-like vehicle that has a pair of wheels and is pedaled by a reclining pilot. The optimized aerodynamic shape allows the bike to travel at higher speeds than some motor-driven vehicles can reach. The AI-driven computer program suggested the optimized shape for the bike. The team behind the project says that the software allows engineers to cut down on the time and guesswork that comes with product design and optimization.

The folks behind the software say that in industry today, many design processes use trial-and-error-based simulation software that can take several hours. The deep learning tech in their software can perform aerodynamic simulations in a few milliseconds.

The team now hopes that with the world record-setting bike design under its belt, it can raise industry interest in its software. The software could be particularly useful in aeronautics, thermal modeling, and automotive industries, but the developers say that the software could work anywhere intensive numerical simulations are performed.