AWS’s Artificial Intelligence/Machine Learning drives transformation for Formula 1 racing

Sports is a fantastic medium to bring the benefits of Machine Learning to life. Currently, we see various sports organisations across the world, embrace AI/ML technologies to bring fans closer to the game and to all the decisions that their teams, and drivers are making.

Until the 1980s, cars were all mechanical. Computers were too large and slow to be used for race cars, so the driver was the only source of ‘data’ for the racing engineer. As good as drivers are at ‘feeling’ the car, it’s difficult for them to remember objective measurements of how the car performed throughout a session when they’re focused on driving.

Where are we today?

With the exception of driving aids incorporated into the car, data is now widely used in all aspects of Formula 1 decision-making. The data sent from the race car is invaluable for identifying performance and obtaining outcomes, from car development to race strategy. Car design is a highly technical frontier that necessitates the collaboration of the world’s greatest computer scientists, racing engineers, and physicists to develop the highest performing and most elite race car possible while adhering to the rules.

Pushing the racing boundaries 

To enhance the viewing experience of fans, AWS partnered with Formula 1 (F1) to help them use advanced cloud technologies like machine learning and cloud computing. Formula 1 is moving a vast majority of its infrastructure from on-premises data centers to AWS and standardizing on AWS’s machine-learning and data-analytics services to accelerate its cloud transformation. Formula 1 is also responsible for the promotion of the FIA Formula One World Championship, a series of auto racing events in 21 countries in which professional drivers race single-seat cars on custom tracks or through city courses in pursuit of the title for World Championship.

AWS worked with Formula 1 to create a next-generation race car that will be used by teams in the 2022 season. The project was performed using Computational Fluid Dynamics (CFD) over 1,150 AWS compute cores to run detailed simulations of over 550 million data points that modelled the impact of one of the car’s aerodynamic wake on another. Using the unmatched scalability of AWS, Formula 1 was able to reduce the average time to run simulations by 80%—from 60 hours down to 12 hours.

“Who would have thought that machine learning would be helping us in improving player health and safety, design faster cars, and solve all these interesting challenges. It will be fascinating to see how ML will solve many complex and important challenges in future. As always, we will continue to work with our customers and help them innovate and solve their biggest opportunities and challenges.” – Priya Ponnapalli, Senior Manager, Amazon Machine Learning (ML) Solutions Lab

The breakthrough between AWS and Formula 1

Formula 1 began working with Amazon Web Services in 2018 to enhance race strategies, data tracking systems, and digital broadcasts through a wide variety of AWS services. These services include Amazon SageMaker, a fully managed machine learning service that enables everyday developers and scientists to easily build and deploy machine learning models, AWS Lambda, AWS’s pioneering event-driven serverless computing service, and AWS analytics services — to uncover never-before-seen metrics that will change the way fans and teams enjoy, experience, and participate in racing.

Through this partnership, AWS’s technology is enabling Formula 1 to improve the way it utilizes data for its business, improve both driver and vehicle performance as well as make future decisions about the look and feel of the car and rules of the race. (Read: AWS Machines Have Evolved Along With The F1 Cars)

Using Amazon SageMaker, Formula 1’s data scientists are training deep-learning models with 65 years of historical race data to extract critical race performance statistics, make race predictions, and give fans insight into the split-second decisions and strategies adopted by teams and drivers. Additionally, in combination with Amazon Kinesis, a product used to collect, process, and analyze streaming data in real-time, SageMaker can detect how a driver is performing mid-ride, determining whether drivers have pushed themselves over the limit.

“We have partnered with Formula 1 on some very key events which includes, transforming the sport, changing how they collect data, how they analyse it and how they visualise it. Our goal is to enhance fan engagement through all the Formula 1 sites, where we’ve partnered on creating a better racing product by helping them develop the next generation of the car. The most challenging aspect is always at the onset and that is the part where we collaborate with Formula 1 experts and start working backwards from there,” says Priya.

The future is bright 

Engineers will progressively rely on machine learning to speed up results and identify design ideas that they would not have come up with otherwise. This methodology is not yet used widely, but it is expanding rapidly across many engineering companies and disciplines, thanks to the enormous potential and major developments in recent years that have addressed its flaws. There are a plethora of areas in which AI can be used to enhance the developments.

AI as a space can use numerous and diverse perspectives and needs all those ideas coming from all the different fields. It’s never been easier to learn a new domain, especially coming up to speed. Today, there are many online courses and great content on ML for people to gain knowledge. So, pull yourself up a chair, bring your unique perspectives and experience to this field.

The post How AWS Adds To The Formula 1 Spectacle appeared first on Kunal's F1 Blog.