Fastest possible lap times through data and analytics

The highest resolution data we are currently modeling comes from sensors on Sportbikes. These sensors are important for capturing a number of different variables including speed, lean angle, brake pressure, suspension performance and exact location on the track for any given rider action. There are up to 22 variables that can be analyzed from bike sensors with data capture reaching hundreds of events per second. When combined with an additional 10 variables representing track configuration and weather, the analyses that can be done to influence setup and rider behavior are amazing. Working with such large amounts of high-resolution data requires adaptive parallel computing and machine learning in order to identify issues with bike setup and rider behavior. Once identified, these issues can be isolated to measure the impact changes have on improving lap times while preserving a margin of safety. Riders and technicians can then prioritize changes that are having a material impact on performance while understanding countervailing issues.

It’s all about going faster and getting back to the paddock on two wheels. The number of factors that go into making each turn’s entry and exit consistent and fast is staggering but can be captured and analyzed through data science. These same principles are portable to other disciplines in motorsports and will be pursued in the future.

Our mission is to provide high resolution analysis showing the strengths and development necessary for optimal machine setup and human decisions to accelerate the safe improvement of lap times.