Wander 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 full of real-time digital feedback from the cars. I’ve been in many of these pits over the years and marveled at the data flows, but I’ve never seen an example of the Microsoft Visual Studio software development suite running right there in the middle of the mess.
But then, I’ve never watched anything like last weekend’s inaugural Abu Dhabi Autonomous Racing League event. A2RL, as it’s known, isn’t the first autonomous racing series: There’s the Roborace series, which saw autonomous race cars set fast lap times while avoiding virtual obstacles, and the Indy Autonomous Challenge, which ran most recently at the Las Vegas Motor . Speedway during CES 2024.
While Roborace focuses on single-car time trials and the Indy Autonomous series focuses on oval action, A2RL set out to break new ground in a few areas.
A2RL put four cars on track, competing simultaneously for the first time. And, perhaps most importantly, it pitted the top performance self-driving car against one man, 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 the teams staffed with an impressively diverse cadre of engineers, ranging from fledgling coders to PhD students to full-time race engineers, all racing to push the limit in a very new way.
Unlike Formula 1, where 10 manufacturers design, develop and produce cars completely to order (sometimes with the help of artificial intelligence), A2RL racing cars are entirely standardized to provide a level playing field. The 550bhp machines, borrowed from the Japanese Super Formula Championship, are identical and teams are not allowed to change a single part.
That includes the sensor array, which has seven cameras, four radar sensors, three lidar sensors, and GPS to boot — all of which are used to perceive the world around them. As I would learn while wandering the pits and talking to the various teams, not everyone takes full advantage of the 15 terabytes of data each car uploads each lap.
Some teams, like Indianapolis-based Code19, started working on the monumental task of creating a self-driving car a few months ago. “There are four groups of startups here,” said Code19 co-founder Oliver Wells. “Everyone else has been competing in races like this, some of them for up to seven years.”
It’s all about the code
Image credits: Tim Stevens
Munich-based TUM and Milan-based Polimove have extensive experience running and winning both Roborace and the Indy Autonomous Challenge. This experience is ongoing, as is the source code.
“On the one hand, the code is constantly being developed and improved anyway,” said Simon Hoffmann, team leader at TUM. The team made adjustments to change the cornering behavior to suit the tighter turns on the road course and also adjust the overtaking aggression. “But overall, I’d say we use the same basic software,” he said.
Through a series of numerous qualifying rounds throughout the weekend, the most experienced teams dominated the schedules. TUM and Polimove were the only two teams to complete lap times under two minutes. Code19’s fastest lap, however, was just over three minutes. the other new groups were much slower.
This has created a competition rarely seen in software development. While there have certainly been competitive coding challenges before, like TopCoder or Google Kick Start, this is a very different thing. Code improvements mean faster lap times — and fewer bugs.
Kenna Edwards is a Code19 assistant race engineer and student at Indiana University. He brought some previous application development experience to the table, but had to learn C++ to write the team’s anti-lock braking system. “It saved us at least a couple of times from crashing,” he said.
Unlike traditional coding problems that might require debuggers or other tools to track down, the improved algorithms here have tangible results. “One nice thing was to see the flat spots on the tire improve over the next session. They have either decreased in size or frequency,” Edwards said.
This application of theory not only creates engineering challenges, but also opens up viable career paths. After previous internships with Chip Ganassi Racing and General Motors, and thanks to her experience with Code19, Edwards starts full-time at GM Motorsports this summer.
A look towards the future
Image Credits: Tim Stevens
This kind of development is a huge part of what A2RL is all about. Shadowing the main on-track action is a secondary series of competitions for younger students and youth teams around the world. Prior to the main A2RL event, these teams competed with stand-alone 1:8 scale model cars.
“The goal is, next year, to keep the smaller car models for the schools, we’ll keep for the universities that maybe do it in the karts, a little bit bigger, they can play with the autonomous karts. And then if you want to be in the big class, you start racing these cars,” said Faisal Al Bannai, secretary-general of the Abu Dhabi Advanced Technology Research Council, ATRC. “I think by seeing this pathway, I think you’re going to encourage more kids 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 numerous teams, some of whom have been testing in Abu Dhabi for months. They also threw a world-class party for the main event, with concerts, drone races and a ridiculous fireworks show.
The action on the track was a little less spectacular. The first attempt at a four-car autonomous race was aborted after one car spun, impeding the following cars. The second race, however, was much more exciting, with a pass for the lead when the Unimore team car of the University of Modena went out. It was TUM who made the pass and won the race, taking the lion’s share of the $2.25 million purse.
When it came to man vs. machine, Daniil Kvyat made quick work of the self-driving car, passing it not once but twice to huge cheers from the assembled crowd of more than 10,000 spectators who took advantage of free tickets to see a bit of history — plus an estimated 600,000 more via event stream.
The technical errors were unfortunate. Still, it was a remarkable event to show just how far autonomy has come — and of course, how much more progress needs to be made. The fastest car was still over 10 seconds off Kvyat’s time. However, he ran smooth, clean laps at impressive speed. This is in stark contrast to the first DARPA Grand Challenge in 2004, which saw each competitor either crash into a dam or meander through the desert on an unplanned sojourn.
For A2RL, the real test will be whether it can grow into a financially viable series. Advertising drives most motorsports, but here, there is the added benefit of developing algorithms and technologies that manufacturers could reasonably apply to 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 he does what he does. This belongs to each of the groups. It doesn’t belong to us.”
The real race, then, may not be on the track, but in securing partnerships with manufacturers. After all, what better way to inspire confidence in your autonomous technology than to show it can handle track traffic at 160 mph?
