Go beyond the Formula One track and see how AI can optimize pit stops, redefine car designs, and make fans happy as the sport's ultimate unseen pit crew member.
Formula One (F1) is one of the most high-tech sports in the world. With cars reaching ear-shattering speeds of 230 mph and making two-second pit stops, it demands the most innovative engineering. F1 is as much about the drivers and cars as it is about the strategies created by the pit crew members analyzing the race.
There are various technologies at play during a race, and AI is becoming one of the most vital tools at the pit crew’s disposal. Let’s take a closer look at where exactly AI is being used in F1.
When F1 drivers go beyond the track's edge with all four wheels, it is considered a track limit violation. The Fédération Internationale de l’Automobile (FIA) verifies these violations, and penalties are handed out based on the results.
Hundreds of violations have to be processed every race. During the 2023 Austrian Grand Prix, only four humans processed around 1,200 potential track limit violations. In the following races, despite increasing the number of people working to check track limit violations, it wasn’t enough.
So, at the 2023 Abu Dhabi Grand Prix, the FIA turned to computer vision. They are using shape analysis to identify the track edge and calculate the number of pixels that extend beyond that line. This layer of the system will be to eliminate cases where human interaction is clearly not needed. It lets the FIA focus on cases that actually need their attention.
Seventy-one percent of sports professionals believe fan engagement is vital to achieving their goals. When people feel an emotional connection to sports, they keep coming back, and it helps the industry generate revenue.
A race weekend involves more than just cheering on your favorite team. Amazon Web Services (AWS) has teamed up with F1 to provide deep insights into split-second decisions and showcase performances with detailed statistics. They are able to do this by analyzing about 70 years of race data stored on Amazon S3. Aside from historical data, machine learning models can analyze the data points collected from over 300 sensors from each F1 car. We are talking about more than 1.1 million data points per second!
It’s only natural that the CEO of Oracle Red Bull Racing, Christian Horner, believes, “Data is in the team’s lifeblood. Every element of performance - how we run a race, how we develop a car, how we select and analyze drivers - it’s all driven by data.” Let’s take a look at some of the statistics these systems are able to output:
Simulation racing, or sim racing, is a virtual F1 racing experience. It is often used to help train drivers to become more familiar with the race track and improve their racing skills without risking injuries or damaging the cars. By incorporating AI into sim racing, teams can simulate dynamic racing conditions, car performance under various setups, and even the behavior of competitors on the track.
Physics engines can accurately model vehicle behavior. They take into account factors like aerodynamics, tire grip, and suspension settings. Meanwhile, data from both real-world races and simulations is constantly analyzed to refine strategies and improve performance. Sim racing setups can range from basic setups with a steering wheel and pedals to full-scale simulators that include motion platforms, VR headsets, and detailed replicas of F1 car cockpits.
Some of the top F1 teams are actively using AI and even have AI companies signed on as their official sponsors. Let’s quickly explore a few of these partnerships and the value they bring.
G42 is a leading AI and cloud computing company based in the UAE. They are the official sponsor for the Mercedes-AMG PETRONAS F1 Team. G42 equips the team with advanced data analysis and machine learning capabilities. With G42’s support, the team can process vast amounts of data in real-time and extract valuable insights to make data-driven decisions. For instance, G42's AI algorithms can analyze telemetry data to optimize car setups for specific tracks, enhancing performance by fine-tuning aerodynamics, tire pressure, and fuel load.
The Red Bull Racing Team uses AI to optimize fuel consumption, helping them to optimize fuel usage, therefore go faster for longer, which can be a critical factor in winning races. This Formula One team won the 2023 Drivers’ Championship and Constructors' Championship in a record-breaking season. The team relies on Oracle Cloud to power race strategy, engine development, simulation racing, fan engagement, and more.
Amazon Web Services (AWS) is one of the official sponsors for the Scuderia Ferrari F1 Team. The Scuderia Ferrari team made a virtual ground speed sensor using (AI) and machine learning through Amazon SageMaker. They were able to deliver quicker and more dependable data to its engineers. The team was able to reduce vehicle weight, which is a critical factor in a sport where even a gram matters. They also used AWS to develop ML models rooted in game theory to analyze the variables in race strategy.
The 2024 season began in March with the Bahrain Grand Prix. We’ve only been through four races so far, but it’s been a thrilling start to the season. Right from the start, we are seeing new AI innovations making their debut for this season.
Let’s start with the push to bring fans closer to the action. It has led to the introduction of new camera angles. The F1 broadcasting team is working closely with Aston Martin to develop a rear light camera. The idea behind it is to give us a view right from the back of the car, capturing the intensity of the race in a way we haven't seen before. AI helps ensure that these images are crisp and clear, adjusting focus and exposure in real-time to deal with the challenges of speed and changing lighting conditions.
With respect to broadcasting, there is also a new revamped replay system that is powered by AI. This AI system can instantly sort through footage to highlight key moments, making sure fans don't miss any of the action. It even has the ability to create slow-motion replays from regular footage, adding a new layer of depth to the viewing experience.
There's also a lot of buzz about the potential use of drones to capture live footage, inspired by a viral First-Person View (FPV) shot of Max Verstappen on a test lap. There are still hurdles regarding safety to overcome. But, the possibility of including drone shots in the future is an exciting one. It's all about finding new ways to bring the thrill of the race to viewers at home.
Speaking of the thrill, the audio part of the broadcast is also set up to get an upgrade. The aim of this upgrade is to make viewers feel like they're right there on the track, with the roar of the engines surrounding them. AI algorithms are being used to fine-tune the audio capture and processing so that the broadcast sound is immersive without being noisy. We want to hear the engines accelerate but still be able to enjoy the race without too much volume.
While AI can be a useful tool, it can’t replace human drivers and pit crew with years of experience and talent. That said, it’s going to be interesting to see how AI will impact Formula One in the future. More advanced technology means more informed decisions on the track, leading to amazing battles for that checkered flag!
Check out our GitHub repository to learn more about AI. Visit our solutions pages to see how AI is being applied in fields like manufacturing and agriculture.
Begin your journey with the future of machine learning
