Episode 3: At Formula 1, a need for speed with machine learning

No sport has been as dynamic in its evolution and embrace of new technology as auto racing. Formula 1®, the commercial rights holder for the auto racing sport, has taken the sport to even higher technical levels with the help of machine learning.

In this episode of the new podcast series, Ahead of the Pack, Tim Crawford speaks with Rob Smedley, Director of Data Systems, Formula 1®, and AWS Technical Ambassador.

Smedley describes his role as taking a “goldmine” of data – vehicle sensor data, aerodynamic data, tire data, timing data – and “packaging [it] into bundles that our audience can understand, which helps to engage our audience and give them a better experience in a Formula 1® race.”

Machine learning dates back several years at Formula 1®, where it was initially used to help the racing teams better understand car performance. Now, Smedley’s team has expanded its focus to fan experience, with many innovative applications that leverage analytics and ML, including F1 Insights powered by AWS and the F1 ProAm DeepRacer event.

“There’s a lot of complexity and massive physics going on in the background,” he said. “But the stuff that pops out on the screen is simple and engaging and brings about that excitement and that tangible entry point into all of these different stories that are going on as a race is unfolding.”

While the use of analytics in Formula 1® is unique, the approach that Smedley’s team takes with data and machine learning is applicable across industries.

“I think in any business, your first port of call is to look at what your problems are,” Smedley said. “And once you understand your problems, then you put in place a strategy for fixing those problems.”