Before we look at how F1 is leveraging artificial intelligence, it’s worth understanding how the AI industry interacts with sport and why.
Alexandre Bonnet, chief engineer of machine learning solutions at Encord, explained to Motorsport.com that the breadth of what F1 offers as a platform for AI and machine learning is attractive to those who want to get involved.
“The reason F1 is so attractive is that there is a huge variety of things you can do with all different types of technology, from the manufacturing stage, like parts, quality analysis , the optimization of industrial processes, robotic manufacturing, even the elements found on cars. Then optimizing vehicle design, things like biomimicry which has come up in recent years, all the way to actual TV broadcasting, that’s, again, a whole other cycle of computer vision and of the AI work that is underway.”
Bonnet also believes that AI companies can benefit from their association with a sporting discipline that has always been a pioneer in the use of technology. “AI companies want to be involved because it’s a complex sport, where you’re at the cutting edge of technology, with the fastest cars, the fastest drivers and everything else, which doesn’t This is not really the case for other well-known sports.”
“F1 teams specialize in making cars and, of course, they have internal AI talent, but they also need to draw on external expertise in the most advanced machine learning systems. recent and most innovative.”
Lewis Hamilton at the 2024 Las Vegas GP.
Photo de: Erik Junius
F1 teams
Formula 1 has always been a data-driven sport: lap times, gaps, top speeds and tire temperatures to name a few. But the amount of information available to a team wanting to analyze the numbers and make the most of the results has until now made the task difficult and tedious.
Today, AI can sort these statistics in seconds, as Andrew McHutchon, McLaren’s head of data science, explained in a recent interview on the team’s website. “Before, we collected data that we didn’t know what to do with, but now with AI and working with Dell Technologies’ AI Factory, we can process the data in a much richer way to extract learnings from it significant.”
“If it’s a pit stop decision, you may only have a third of a lap before the car goes into the pit lane, and after that your luck has passed, so you have to be quick. You might have terabytes and terabytes to analyze, and it would take half a day or more to answer a single question without AI.”
In Formula 1, it’s all about being the fastest, and this speed of the AI is useful off the track as well as on the track. “Even when it comes to the team at the factory working on developing the car, speed is important”added McHutchon.
Charles Leclerc
Photo de: Sam Bloxham / Motorsport Images
“You can have five questions, and if it takes you half a day to answer each one, everything slows down. AI speeds all that up, and the faster we can answer those questions, the faster we can develop the car and the more chances we have to win championships.”
F1 fans
AI will therefore undoubtedly play a role in the races of the future, but what about for spectators? AWS has been a technology partner of Formula 1 since 2018 and launched “F1 Insights” a year later.
Using just a fraction of the million data points provided by each car, every second, there are now 23 “F1 Insights” infographics available to TV producers, including forecasts on pit strategy, number of laps before overtaking or undercuts.
This information is used to improve the viewer experience, as explained by Neil Ralph, senior manager of sports partnerships at AWS. “Drive To Survive has attracted a large number of new fans to Formula 1 and they are trying to understand what a race is like. We are using data from the sport to unpack the complexities of this spectrum of the fan base.”he told Motorsport.com.
“By comparison, it is much more difficult to understand the inner workings of an F1 race than, say, a football match where much of the attention is on one area of the pitch, whereas F1 has a 5K track with 20 cars spread across its length, so you can only cover part of the ground with video.”
The start of the 2024 Singapore GP.
Photo by: Zak Mauger / Motorsport Images
“So using data to tell stories helps and increases engagement for new fans, while enriching the experience for long-time fans. AI has allowed us to dissect racing strategies, windows of pit stops and undercuts; using these data-driven on-screen graphics, we can help every fan understand what’s happening and allow commentators around the world to keep the excitement level up.”
David Croft, lead commentator for Sky Sports F1 since 2012, is one of the professionals who now have the ability to use the information provided by AWS. “For me, AI is a great help for commentators and audiences because it helps understand how a story is developing, but I don’t want it to tell people how the story is going to end.”he explained to Motorsport.com.
“Live sport is unpredictable, joyful and heartbreaking all at the same time, depending on who you’re following, so as long as it doesn’t ruin the outcome, what’s on offer is fantastic.”he added. “It doesn’t necessarily have a huge impact on my work: I comment, I’m in the moment, I absorb the data and I relay it to tell the story.”
-“There are limits to what I can watch! Where I think it’s really useful is in situations like, in the past, when Martin [Brundle] and I estimated the moment when we thought one driver would be right behind the other.”
The start of the 2024 Belgian GP.
Photo de: Michael Potts / Motorsport Images
“Today this information is out there and everyone knows it, thanks to real-time data and AI simulations, and more often than not it’s very accurate. Plus, when you say ‘the moment a driver will catch another will be in four laps’, that doesn’t mean the story is over. It means we have to stay focused and start getting excited so we can start to intensify the storytelling.
Croft may be worried that AI will go too far in predicting Formula 1, but how far can it go in the viewer experience? Ralph has his own idea on the matter. “We don’t see all 23 infographics every weekend, but as we talk about personalization and giving fans a choice on what they see, maybe in the future it won’t be the team The F1 technical production team will decide which graphics will be available – the graphics will be available to everyone and you will decide what you want to see.”
“Viewers will be able to choose how much data they want to see, perhaps through a second screen, giving them a choice between following the cars or having a large amount of data.”
F1 drivers
With the public sated and the design, performance and strategy of the cars settled, it seems the only thing in Formula 1 that will remain sacrosanct in the AI era is the role of the driver… or not ?
When Motorsport.com raised this question with Kevin Magnussen of Haas, he replied: “Really? If you’re just looking for efficiency, the AI will end up driving the car better than us, there’s no doubt about that. But it’s not really entertainment anymore, I think it would really be boring to watch a Formula 1 race driven by computers, that wouldn’t interest me.”
Kevin Magnussen, Haas F1 Team
Photo by: Simon Galloway / Motorsport Images
“You can identify with the driver of the car and what he’s going through, the ability of a human being to drive an F1 car is fascinating because you can identify with him. When I watch footballers, I find them incredible; I know how good I am at kicking a ball and the level they reach is fascinating and I think this relationship must exist, otherwise there is no point.”
With the help of AI, teams are developing new and faster ways to analyze data that could prove crucial to the outcome of a race weekend. All these statistics can be fed into as many software programs as possible, but all results obtained ultimately end up back at the driver behind the wheel.
Far from the racing of yesteryear, the new way to approach marginal gains is to use AI to facilitate strategic decisions and pit calls, which Magnussen believes can only grow over time. “These things are already in effect in some way, but it’s done manually, by running simulations on millions of different scenarios and suggesting a strategy.”
“In the future your strategist might be a computer, and further still you might see AI tuning your car and developing your aerodynamics, but if you take away the human element, people watching you won’t find it interesting .”
In this article
Mark Mann-Bryans
Formula 1
Kevin Magnussen
Be the first informed and subscribe to email alerts to receive information in real time
Subscribe to news alerts