Enhancing Pit Stop Performance with Machine Learning: Scuderia Ferrari HP’s Partnership with AWS
Revolutionizing Data Analysis in Formula 1® Racing
Challenges of Traditional Pit Stop Performance Analysis
Modernizing Pit Stop Techniques Through Innovative Cloud Solutions
Driving Innovation in Racing Technology Together
Developing a Next-Generation ML-Powered Pit Stop Analysis Solution
Conclusion: Achieving Record-Breaking Pit Stops with Precision and Speed
Optimizing Speed: How AWS is Revolutionizing Pit Stop Analysis for Scuderia Ferrari HP
Formula 1 (F1) is renowned for its speed and precision, where every millisecond counts—not just on the track but also during pit stops. As drivers race against time to change tires or address any damages, the efficiency of their pit crew can mean the difference between victory and defeat. Scuderia Ferrari HP is harnessing the power of Amazon Web Services (AWS) and machine learning (ML) to enhance pit stop analysis, transforming a challenging process into a streamlined operation.
Challenges with Pit Stop Performance Analysis
Historically, analyzing the performance during pit stops has been an arduous task. Track operations engineers had to painstakingly sift through hours of footage, reviewing videos from multiple angles while correlating them with telemetry data—from tire changes to car diagnostics. Over a typical race weekend alone, engineers would encounter an average of 22 videos per driver, leading to a staggering 600 videos per season.
This manual review process is not only time-consuming, but it also carries the risk of inaccuracies. With AWS’s new solution, engineers can now synchronize data up to 80% faster, drastically reducing the time spent analyzing pit stops.
Modernizing Through Partnership with AWS
The collaboration with AWS is a game-changer for Scuderia Ferrari HP as it modernizes the labor-intensive task of pit stop analysis. The introduction of cloud technology and ML allows for unparalleled precision and speed.
Marco Gaudino, Digital Transformation Racing Application Architect, states, “Previously, we had to manually analyze multiple video recordings and telemetry data separately, making it difficult to identify inefficiencies. With this new approach, we can now automate and centralize the analysis, gaining a clearer and more immediate understanding of every pit stop.”
Through the use of object detection in the AWS SageMaker AI platform, video captures are synchronized with telemetry data. This innovative solution leverages a serverless architecture that optimizes compute resource usage, helping Scuderia Ferrari HP remain compliant with the strict budget caps from the FIA.
Driving Innovation Together
Since 2021, AWS has been an integral partner for Scuderia Ferrari HP, not just on the track but off it as well. Collaborations have led to innovations like a virtual ground speed sensor to reduce vehicle weight and enhancements in the power unit assembly process.
The pit stop analysis solution was first tested in March 2024 at the Australian Grand Prix, moving into production at the following Japanese Grand Prix. It quickly became essential in providing Scuderia Ferrari HP a competitive edge.
Furthermore, the team is working on a prototype to automatically detect anomalies during pit stops, such as issues with tire installation or lifting delays. In preparation for the 2025 season, they plan to enhance their camera setup to capture higher frame rates, allowing for even more detailed analysis.
Developing the ML-Powered Pit Stop Analysis Solution
The cutting-edge ML solution efficiently correlates video and telemetry data. Using the YOLO v8 algorithm, it identifies critical signals, such as the green light that indicates when to leave the pit. The system allows engineers to review synchronized footage using a custom visualization tool.
Gaudino emphasizes, “By systematically reviewing every pit stop, we can identify patterns and refine our processes. This leads to greater consistency and reliability, reducing the risk of errors.”
The process workflow involves several AWS components:
- Video Upload: When a pit stop occurs, front and rear videos are automatically transferred to Amazon S3.
- Event Trigger: Amazon EventBridge connects multiple Lambda functions to process the video.
- Data Synchronization: Lambda retrieves timestamps and merges videos with telemetry to achieve a synchronized output.
Previously, manually correlating this data could take minutes; now it’s down to 60-90 seconds, producing near-real-time insights.
Conclusion
Thanks to the new ML-powered pit stop analysis solution, Scuderia Ferrari HP has successfully streamlined its analysis process, achieving the fastest pit stop in each race of the 2025 season, with a record time of just 2 seconds in Saudi Arabia for Charles Leclerc. By leveraging AWS technologies, the team is focused on getting its drivers back on the track more efficiently, ultimately aiming for the best results on race day.
To dive deeper into how machine learning and cloud solutions can transform your operations, check out Getting Started with Amazon SageMaker AI and discover more about Scuderia Ferrari HP’s use of AWS services.
About the Author
Alessio Ludovici is a Solutions Architect at AWS, dedicated to driving innovation through advanced technology solutions.