Explore the pits at any professional motorsports event, particularly something like Formula 1, and you’ll encounter numerous computer displays filled with telemetry. Modern teams are immersed in real-time digital feedback from the cars. I’ve had the opportunity to visit many of these pits over the years and have been amazed by the constant streams of data. However, I have never witnessed the presence of the Microsoft Visual Studio software development suite in the midst of such chaos.
The inaugural Abu Dhabi Autonomous Racing League event, known as A2RL, offered a unique experience. Unlike other autonomous racing series like the Roborace series and the Indy Autonomous Challenge, A2RL set out to break new ground by having four cars on track simultaneously for the first time. Additionally, the event featured a showdown between the top-performing autonomous car and former Formula 1 driver Daniil Kvyat.
The real challenge of the event lay behind the scenes, where teams composed of a diverse group of engineers, from novice coders to doctoral students to experienced race engineers, were striving to push the limits in a new way.
Unlike Formula 1, where manufacturers design and produce bespoke cars, the A2RL race cars are standardized to ensure fair competition. These 550-horsepower machines, borrowed from the Japanese Super Formula Championship, are identical, with teams prohibited from making any modifications.
The sensor array on the cars includes seven cameras, four radar sensors, three lidar sensors, and GPS, all utilized to perceive the surrounding environment. Despite the vast amount of data collected by each car per lap, not all teams are fully leveraging this information.
Some teams, like Code19 from Indianapolis, only started working on the project of creating a self-driving car a few months ago. Code19 co-founder Oliver Wells noted, “There’s four rookie teams here. Everyone else has been competing in competitions like this, some for up to seven years.”
The Significance of Coding
Munich-based TUM and Milan-based Polimove, with extensive experience in Roborace and the Indy Autonomous Challenge, dominated the event with their developed source code that carried over from previous competitions.
“On the one hand, the code is continuously developed and improved anyway,” said Simon Hoffmann, team principal at TUM. The team made adjustments to enhance cornering behavior and overtaking aggression to suit the road course’s sharp turns. “But in general, I would say we use the same base software,” he mentioned.
Throughout the qualifying rounds, experienced teams like TUM and Polimove outperformed newer teams, with only these two teams managing lap times under two minutes. In contrast, Code19’s fastest lap was slightly over three minutes, showcasing the performance gap.
This competition presents a unique challenge in software development, where improvements in code directly impact lap times and safety on the track.
Kenna Edwards, a Code19 assistant race engineer and student at Indiana University, shared her experience of learning C++ to develop the team’s antilock braking system, which prevented potential crashes during the event.
Unlike traditional coding challenges that may require debuggers, improvements in algorithms in this scenario yield visible results, such as improved tire performance over sessions. Edwards reflected, “A cool thing has been seeing the flat spots on the tire improve over the next session, either they’ve reduced in size or in frequency.”
Implementing theory in practice not only presents engaging engineering challenges but also opens up potential career opportunities. Edwards, with internships at Chip Ganassi Racing and General Motors, will start full-time at GM Motorsports this summer.
Focusing on the Future
The A2RL event aims to inspire the next generation of engineers and innovators through secondary competitions for younger students and youth groups, allowing them to engage with autonomous model cars.
Faisal Al Bannai, the secretary general of Abu Dhabi’s Advanced Technology Research Council (ATRC), expressed plans to expand the scope of the competition to include larger-scale autonomous go-karts for universities, creating a pathway for aspiring competitors to progress from small-scale models to full-sized autonomous race cars.
ATRC’s sponsorship of the event covers various expenses, from cars to team accommodations, fostering a conducive environment for innovation and competition. The on-track action, while facing technical challenges, showcased the advancements in autonomy and highlighted the potential for further progress.
The ultimate test for A2RL will be its ability to evolve into a sustainable series. While traditional motorsports rely on advertising, this event offers the added benefit of developing algorithms and technologies that could be implemented by manufacturers.
According to Al Bannai, teams retain ownership of the code developed for the event, potentially leading to licensing opportunities for their AI algorithms in the future.
The event demonstrated the evolving landscape of autonomy and the potential applications of autonomous technology in real-world scenarios. The blend of competition and innovation in A2RL signifies a step towards shaping the future of autonomous racing and technology.
