Artificial Intelligence is an effective tool for making a faster bike. With software, engineers can calculate the most aerodynamic shape for a bicycle. The air resistance must be as low as possible to optimize the speed.
AI for faster bikes
A machine learning algorithm has been trained to recognize the aerodynamic properties of various 3D shapes. The algorithm predicts the aerodynamic performance from previous experiences, as an engineer would also tackle it. This makes it possible to estimate a new design much more quickly. Hours become miliseconds.
The algorithm is now used in the development of a new fast ‘aero-speed bike’. This bike will compete for the title at ‘World Human-Powered Speeds Challenge’ in Nevada in September.
World Human Powered Speed Challenge
The competition consists of university teams with students. The bicycle of the world record holder, Sebastiaan Bouwier, already seems optimized to the maximum. The maximum speed is 133.78 kilometers per hour. Nevertheless, an AI system can make a bicycle 5 to 20 percent more aerodynamic. The bike looks like a small racing car and is made so that the cyclist fits perfectly.
The software can also be used for other areas, such as the design of drones, wind turbines and aircraft.