The very first race between four autonomous cars and a Formula 1 driver has just taken place in Abu Dhabi

Walk through the pits at any professional motorsport event, especially something like Formula 1, and you’ll see endless computer screens full of telemetry. Modern teams are inundated with real-time digital feedback from cars. I’ve visited many of these pits over the years and marveled at the data streams, but I’ve never seen an instance of the Microsoft Visual Studio software development suite running there in the middle of the chaos.

But then, I’ve never been to anything like the inaugural Abu Dhabi Autonomous Racing League event last weekend. A2RL, as it’s called, isn’t the first autonomous racing series: there’s the Roborace series, which saw autonomous racing cars achieve fast lap times while avoiding virtual obstacles; and the Indy Autonomous Challenge, which recently took place at Las Vegas Motor Speedway during CES 2024.

While the Roborace focused on single-car time trials and the Indy Autonomous Series focused on oval action, the A2RL set out to innovate in a few areas.

A2RL put four cars on track, competing simultaneously for the first time. And, perhaps most importantly, it pitted the most capable self-driving car against a human being, former Formula 1 driver Daniil Kvyat, who drove for various teams between 2014 and 2020.

Image credits: Autonomous Racing League

The real challenge was behind the scenes, with teams of impressively diverse engineers, from entry-level coders to PhD students to full-time racing engineers, all struggling to find the limit of an entirely new manner.

Unlike Formula 1, where 10 manufacturers design, develop and produce completely bespoke cars (sometimes with the help of AI), A2RL race cars are completely standardized to provide a level playing field. The 550 horsepower machines, borrowed from the Japanese Super Formula championship, are identical, and teams are not allowed to change a single component.

This includes the sensor network, which includes seven cameras, four radar sensors, three lidar sensors and a GPS to boot, all of which are used to perceive the world around them. As I would learn from walking around the pits and talking to the various teams, not everyone makes full use of the 15 terabytes of data that each car sucks up every lap.

Some teams, like Indianapolis-based Code 19, only started working on the monumental project of creating a self-driving car a few months ago. “There are four rookie teams here,” said Oliver Wells, co-founder of Code 19. “Everyone has been doing competitions like this, some for seven years.”

It’s all about code

Image credits:Tim Stevens

Munich-based TUM and Milan-based Polimove have extensive experience racing and winning the Roborace and Indy Autonomous Challenge. This experience lives on, as does the source code.

“On the one hand, the code is continually developed and improved,” said Simon Hoffmann, team leader at TUM. The team made adjustments to change the cornering behavior to suit the tighter bends in the road and also to adjust the aggressiveness of overtaking. “But in general, I would say we use the same basic software,” he said.

Throughout the numerous qualifying rounds throughout the weekend, the more experienced teams dominated the timing. TUM and Polimove were the only two teams to achieve lap times under two minutes. However, Code 19’s fastest lap lasted just over three minutes; the other new teams were much slower.

This has created competition rarely seen in software development. While there have certainly been competitive coding challenges, like TopCoder or Google Kick Start, this is a very different thing. Improvements to the code mean faster lap times and fewer crashes.

Kenna Edwards is a Code 19 Assistant Race Engineer and student at Indiana University. She brought some application development experience, but had to learn C++ to write the team’s anti-lock braking system. “It saved us from crashing at least twice,” she said.

Unlike traditional coding problems that may require monitoring by debuggers or other tools, the algorithms improved here yield tangible results. “The cool thing is seeing the punctures on the tires improve over the next session. Either their size or frequency has decreased,” Edwards said.

This implementation of theory not only creates interesting engineering challenges, but also opens up viable career prospects. After previous internships at Chip Ganassi Racing and General Motors, and with his experience with Code 19, Edwards starts full-time at GM Motorsports this summer.

A look to the future

Image credits:

This type of development represents a large part of the very essence of A2RL. The main track action is followed by a secondary series of competitions for young students and youth groups from around the world. Before the A2RL main event, these groups competed with 1:8 scale self-driving model cars.

“The goal is that next year we will keep the smaller car models for schools, we will keep for universities maybe do it on go-karts, a little bigger, they can play with karts autonomous. And then, if you want to get into the big leagues, you start racing these cars,” said Faisal Al Bannai, secretary general of the Abu Dhabi Advanced Technology Research Council, ATRC. “I think seeing that path, I think you’ll encourage more guys to get into research, get into science.”

It is Al Bannai’s ATRC that foots the bill for A2RL, covering everything from cars to hotels for the many teams, some of whom have been testing in Abu Dhabi for months. They also threw a world-class party for the main event, complete with concerts, drone races, and a ridiculous fireworks show.

The action on track was a little less spectacular. The first attempt at a four-car standalone race was aborted after one car spun, blocking following cars. The second race, however, was much more exciting, with a pass for the lead when the Unimore team car from the University of Modena spun out. It was TUM who made the pass and won the race, taking home the lion’s share of the $2.25 million purse.

As for man versus machine, Daniil Kvyat made quick work of the self-driving car, overtaking it not once but twice to immense cheers from the assembled crowd of more than 10,000 spectators who took advantage of free tickets to come see some history — plus about 600,000 other people streaming the event.

The technical problems were unfortunate. It is nevertheless a remarkable event which illustrates how far we have come in terms of autonomy and, of course, the progress that remains to be made. The fastest car was still more than 10 seconds behind Kvyat’s time. However, he delivered smooth, clean laps at an impressive speed. This stands in stark contrast to the first DARPA Grand Challenge in 2004, in which every competitor crashed into a barrier or got lost in the desert on an unplanned trip.

For A2RL, the real test will be whether it can evolve into a financially viable series. Advertising is the driving force behind most motorsports, but here there is the added benefit of developing algorithms and technologies that manufacturers could reasonably apply in their cars.

ATRC’s Al Bannai told me that while the series organizers own the cars, the teams own the code and are free to license it: “What they’re competing on right now is the algorithm, the AI ​​algorithm that makes this car do what it does. This is up to each team. It doesn’t belong to us.

The real race therefore perhaps does not take place on the track, but in the conclusion of partnerships with manufacturers. After all, what better way to instill confidence in your autonomous technology than by showing that it can handle traffic on a race track at 160 mph?