Tech | Visa | Scholarship/School | Info Place

How Neural Concept’s aerodynamic AI is shaping Formula 1 racing

this is a long From pedal bikes to Formula 1 cars. But that’s exactly the quantum leap that AI-based startup Neural Concept and its co-founder and CEO Pierre Baqué have made in just six years.

In 2018, the company’s fledgling software helped develop the world’s most aerodynamic bike. Today, four out of ten Formula 1 teams use a modified version of the same technology.

Along the way, Baqué’s company signed contracts with aerospace suppliers such as Airbus and Safran and secured $9.1 million in Series A funding in 2022. Switzerland-based Neural Concept, which currently has 50 employees and is working on raising Series B funding, has also provided software to historic F1 to help teams like Williams find a way back to the pinnacle of the world’s top motorsport.

However, Formula One cars relied on 1,000-horsepower hybrid V6 engines, and Barker’s first practical application of the technology was human-powered.

pedal power

In 2018, Buck studied at the Computer Vision Laboratory of Ecole Polytechnique Fédérale de Lausanne, working on applying machine learning technology to three-dimensional problems.

“I got in touch with the guy who was leading the team, who was designing a sixth or seventh generation bike, and their goal was to break the world record for bike speed,” Barker said. That person is Guillaume DeFrance and the team is IUT Annecy from Savoie Mont Blanc. The cycling team has gone through six bike design iterations.

“When I came back to him two days later, he almost looked like the current world record holder in shape,” Barker said. The team was impressed and asked for more iterations. The result, according to Baqué, is “the most aerodynamic bike in the world right now.”

It’s a strong statement, but it’s backed up by several world records set in 2019. We’re not talking airfoiled downtubes or dimple edges to reduce drag. The bike is fully shaded and the rider sweats in a composite cocoon, completely sheltered from the wind.

The core technology is a product called Neural Concept Shape (NCS). It is a machine learning-based system that makes aerodynamic suggestions and recommendations. It applies to the broad field of Computational Fluid Dynamics (CFD), where trained engineers use advanced software suites to run three-dimensional aerodynamic simulations.

CFD is much faster than sculpting a physical model and putting it into a wind tunnel. Nonetheless, it is also highly system-intensive and relies heavily on humans to make the right decisions.

At its core, NCS helps engineers avoid potential aerodynamic pitfalls while pushing them in directions they may not have considered. For example, in “co-pilot mode,” engineers can upload existing 3D shapes to provide a starting point.

NCS will then drill down into its neural network to suggest improvements or modifications, as well as possible paths in a choose-your-own-adventure 3D game. Human engineers then single out the most promising suggestions and run them through further testing and refinement, iterating toward aerodynamic glory.

Not just “cheating”

NCS is not only suitable for racing, but also for the automotive and aerospace industries. “The path to widespread adoption in these types of companies is slow,” Buck said of working in the somewhat conservative aerospace industry. “That’s how we start working more with the automotive industry, where the needs are more urgent and things are going to change very quickly.”

Neural Concept has contracts with several global suppliers including Bosch and Mahle. Aerodynamics are becoming increasingly important in the automotive world, and manufacturers are looking for more aerodynamic vehicles to provide the greatest possible range for a given size of battery pack.

But it’s not all about tricking the wind. NCS is also used to develop products such as battery cooling plates, which, if more efficient, can keep batteries at optimal temperatures without using too much energy in the process. “We can achieve huge results,” meaning greater scope, Buck said.

While the ultimate proving ground for these technologies is always the road, the ultimate laboratory is Formula One racing. F1 has been a global motorsport phenomenon since 1950 and is currently experiencing an unprecedented wave of popularity.

The power of Netflix

The Netflix series Formula 1: Drive to Survive brings the excitement of Formula 1 to a whole new audience. While the series focuses on inter-team politics and drama, on-track success has more to do with aerodynamics. This is where the neural concept comes in.

Barker started watching Formula 1 before Reed Hastings even had access to Netflix. “I’ve been watching since the days of David Coulthard and Michael Schumacher.”

Today, parts developed with the help of his company’s software are running at the pinnacle of motorsport around the world. “It’s a big, big feeling of accomplishment,” Barker said. “When I started the company, I thought it was a milestone. Not just the Formula 1 cars, but the parts designed using the software. Yeah, every time that happens, it feels great.”

Formula One is also an extremely secretive sport. Of the four teams Neural Concept works with, only one is willing to be identified as a customer, and even remains tight-lipped about the entire process.

Williams is one of the most legendary teams in Formula One. Founded in 1977 by racing legend Frank Williams, his team dominated the 1990s, winning five Constructors’ World Championships, including three consecutive titles from 1992 to 1994.

But like most sports, success for Formula One teams comes in cycles, and right now, Williams is in a rebuilding phase. The team finished last in the 2022 season and only rose to seventh place last year.

NCS is one of the tools helping Williams regain its competitive edge. “We are using this technology in a number of ways, some of which improve our simulations, while other methods we are working on will help deliver better results for the first time in CFD,” said Williams Head of Aerodynamics Technology Hari Roberts said.

Likewise, CFD simulations are time-consuming and expensive, and the situation is further complicated by Formula 1 regulations that limit teams’ testing capabilities. Physics time in the wind tunnel is strictly limited, and each team has a limited budget for computing time to develop the car.

Any tool that helps a team complete an aerodynamic design quickly is a potential advantage, and NCS is really fast. Baqué estimates that a full CFD simulation that would normally take an hour passes through NCS in just 20 seconds.

And because NCS doesn’t run actual physics-based calculations but instead makes AI-driven guesses based on its aerodynamic learning network, it’s largely free from the strict constraints of F1. “Anything we can do that allows us to gain more knowledge and gain more performance from every CFD and wind tunnel run gives us a competitive advantage,” Roberts said.

But the team still has to pay the price. NCS costs vary depending on team size and type of access, but typically range from €100,000 to €1 million per year, Baqué said. This is a significant commitment considering that annual operating costs for F1 teams are capped at $135 million.

Williams’ Roberts would not point to any specific component or lap time improvements brought about by the NCS software, but said it affected their car’s performance: “This technology is used as a tool for developing our car’s aerodynamics toolset. part. So we can’t directly attribute the lap times to it, but we know it helps our correlation and the speed with which we study the new aerodynamic conditions.”

Beyond aerodynamics

The continued advancement of artificial intelligence will not stop here. There are rumors that human agents on the pit wall could call the shots on race strategy and even car setup.

“This is a fascinating time because the AI/ML industry is growing exponentially,” Roberts said. “However, this is also the real challenge facing anyone involved in technology today. What new tools are we spending our time exploring, developing and adopting?”

It’s not the kind of intrigue that would appeal to the average “survival” viewer, but for many F1 fans, the racing behind the scenes is the ultimate source of drama.

As for Neural Concept, the company is continuing to delve into non-racing areas of the automotive industry, working on developing more efficient electric motors, optimizing cabin heating and cooling, and even entering crash testing.

Buck said the company’s software can help engineers optimize a car’s crashworthiness while reducing unnecessary weight. But currently the company can only perform crash simulations on individual components, not the entire vehicle. “This is one of the few applications where we hit the limits of performance,” he said.

maybe another app EU’s booming AI supercomputing platform?

#Neural #Concepts #aerodynamic #shaping #Formula #racing

Leave a Reply

Your email address will not be published. Required fields are marked *

Index