How AI Could Change Formula One

Over the course of seventy years, Formula One has established itself as a frontier of technological innovation. Year after year, the industry has pushed the boundary of scientific discovery and has pioneered technology after technology after technology. In such an ever-changing sport there is always something new on the horizon and the ‘next big thing’ might just be artificial intelligence.

Since its inception, decision-making has been a critical aspect of the sport. Every race, teams are faced with the challenge of formulating strategies to maximise chances of success. How aggressively should their driver race? What driving mode should the car be set to? When should the driver to pit stop? Traditionally, these decisions are made by race engineers who interpret real-time data and conclude the best course of action. However, when millions of dollars hang in the balance and milliseconds separate victory and loss, there exists virtually no margin for error- and if we know anything about human nature, there is always a margin for error…

This is where AI could be a game-changer. Instead of relying on human input, which can be clouded by various external factors, artificial intelligence could conduct computer analysis of racing situations to make data-backed decisions in real-time. Modern F1 cars all have over 200 sensors which act as data collection systems, constantly monitoring the condition of the vehicle and feeding back terabytes of information to race engineers. But there is one major issue. The sheer amount of this information means humans simply can’t take full advantage of it. However, with the advent of AI, machine learning algorithms could be employed to dissect all of these vast data streams within seconds and conclude the response most appropriate for the situation.

Even off the racetrack, AI could help to propel the sport to new heights. During manufacturing and development, artificial intelligence could be used to analyse the performance of individual components, helping identify weaknesses within the build. This would revolutionise production within the entire automotive industry and also provide smaller teams with less research budget to close the gap with the top dogs. If this were to be realised, it would almost certainly become an indispensable tool for every manufacturer in Formula One, allowing them to build more competitive cars.

Potentially the most significant application of AI would be its contribution to safety. Even though F1 has never been safer, as with any motorsport, drivers are still risking their lives every time they drive out onto the tarmac. Tackling the threat of collisions, AI systems could be programmed to provide warnings when the vehicle is in dangerously poor condition. These warnings would stem from an analysis of various factors that affect the likelihood of an accident- tyre wear, track precipitation, brake temperature etc. On the track, where the margins can be so fine, this intelligent safety monitoring could make the difference that saves a life.  

 So, when can we expect AI in Formula One? Well, a few teams have already taken major strides to incorporate AI into the development of their cars with the main two being Renault and Williams F1. Renault has been partnered with Microsoft since 2017 with the intention of using AI technology to get an edge on other teams. Microsoft stated, ‘the Renault team uses everything from Azure Machine Learning, to Stream Analytics, to Dynamics 365, to help it win’. Additionally, the software giant explained that Renault’s design team was using HoloLens, an augmented reality platform, to assist in component designing. Elsewhere, over 3 years ago, Williams CIO Graeme Hackland asserted that “I have rather wisely or unwisely decided that [by] 2020, the driver will be called in to pit by an artificial intelligence.” This reveals that, whilst progress in the realm has been slightly underwhelming, the future of the sport may very well lie in AI.

 So then, will AI ever truly break into the sport of Formula One? The obvious answer is that no-one knows. Whilst there is much excitement about its huge potential, the idea is often met by sceptics highlighting its impracticality- but one thing is for certain, the future looks bright.

Thumbnail Credit: Spencer Davis