Press & Media
PRESS RELEASE | PR001
Rapson completes maiden A2RL race and debuts its reinforcement-learning AI driver at Yas Marina Circuit
04 December 2025 · Budapest, Hungary
Budapest-based deep-tech racing team Rapson has successfully completed its first race weekend in the Abu Dhabi Autonomous Racing League (A2RL), debuting its in-house-developed AI driver. A2RL is the world’s largest driverless racing series, where teams race modified Super Formula cars – the second most powerful single-seaters after Formula 1 – around Yas Marina Circuit in Abu Dhabi.
Beyond its competitive debut, the Hungarian engineering team also achieved a significant technical milestone: Rapson is the first in the world to entrust a high-performance race car on a real circuit to a reinforcement-learning-based AI driver, outside of simulation.
Rapson’s participation in A2RL is supported by partners such as the Bosch Group in Hungary, GanzKK and Riptides.

Quick Download

Media Kit
High-resolution photos, team logos, and brand assets for editorial use

Media Contact
Rapson Communications
Phone: +36 30 080 4633
Rapson enters A2RL
The Abu Dhabi Autonomous Racing League (A2RL) is an international championship created by ASPIRE, part of Abu Dhabi’s Advanced Technology Research Council (ATRC). A2RL is a professional driverless motorsport series that fields fully autonomous Super Formula race cars, capable of exceeding 280 km/h. Beyond the racing itself, the championship is designed both as a platform to accelerate research in AI, robotics and advanced vehicle-control technologies, and as a way to showcase the future of mobility and its possibilities to the widest possible audience.
“I often hear the doubts about autonomous racing: in traditional motorsport we cheer for the drivers – their courage and personality are what give the sport its soul – and at first glance that seems to disappear when there is no human in the cockpit. I don’t think we should be looking at autonomous racing today as if it were a one-to-one replacement for a human-driver series. The real value here is that the technology is forced to develop in a racing environment, and later it can make everyday mobility safer and more comfortable.
And at the same time, it looks spectacular – I honestly couldn’t imagine a better stage to demonstrate the progress of cutting-edge technology and to promote science and engineering. The series is still in its infancy, which is why we’re especially happy to already be here and, together with the other teams and the organisers, help shape the future of autonomous racing,”
says Ármin Bogár-Németh, CEO of Rapson
In 2025, for A2RL’s second season, 11 racing teams from all around the world gathered at Yas Marina Circuit in Abu Dhabi. Rapson, built around a Hungarian engineering team, competed alongside entrants from Italy, Germany, France, the United Arab Emirates, Singapore, China, Japan and the United States.
“We are proud to take part in AI-based developments that will form the basis of the revolutionised driving of the future – and to do all this together with Hungary’s finest team,”
highlights Gergő Szabó, CEO of GanzKK
World-first step: reinforcement-learning control on a Super Formula race car
During race preparation, Rapson installed its own in-house reinforcement-learning (RL) controller on its A2RL race car. Under this RL-based control, the car performed controlled, low-speed manoeuvres on the circuit. To the team’s knowledge, this was the first time globally that a reinforcement-learning controller had been applied to such a high-performance single-seater formula car in the physical world; until now, experiments of this kind had essentially been limited to simulation.
Reinforcement learning is a self-learning AI technique in which the system improves itself in simulation: it tries different actions in different situations and receives positive or negative feedback (rewards or penalties) based on the outcomes, gradually converging on better and better strategies. In Rapson’s case, the AI “driver” completed millions of training episodes in simulation before ever touching the real car, effectively learning how to control a race car on its own. The milestone lies in the fact that Rapson then took this trained AI driver out of the virtual world, put it into a real race car, and actually handed over control on a live circuit.
“Reinforcement learning is rarely found in real-world scenarios.
Deploying an RL policy on a 500+ horsepower, million-dollar formula car comes with profound challenges; naturally, the wise approach is to start safe and slow.”
explains Gergely Bári, CTO of Rapson
The significance of RL is underlined by the fact that, in recent years, it has been the technique behind several “superhuman” performances, mostly in virtual domains – from AlphaGo and AlphaStar to GT Sophy. Rapson believes that, in the physical world, after drones, high-performance racing – and, through it, road traffic – can be the next domain where this capability truly comes to life.
This first on-car demonstration is therefore a key step in Rapson’s long-term vision: to develop superhuman motion intelligence – AI with proven race-pace capabilities – that can gradually be transferred from the racetrack into everyday road vehicles.
“The development of autonomous and intelligent vehicle systems is one of the Bosch Group’s strategic focus areas: we believe that the proper use of artificial intelligence and advanced sensor technology can make mobility safer, more predictable and more comfortable for people. The international performance of this Hungarian engineering team, the Rapson project, clearly shows how research can turn into real, working solutions. We are convinced that developments like this bring us closer to a future in which technology truly works for people, in line with the principle Bosch has represented for decades: "Invented for life".”
emphasizes Dr. István Szászi, representative of the Bosch Group in Hungary and the Adriatic region
Road to the race
The weeks of preparation and qualification before the race presented Rapson’s engineers with a wide range of challenges. They had to get to know the car as a race car, understand the behaviour of each subsystem, layer their own decision-making and control software on top of the baseline stack provided by A2RL, and then optimise it all for the unique layout and conditions of Yas Marina Circuit
To qualify for the main event, teams had to meet three criteria:1. respond correctly to race-control flag signals. 2. achieve a lap time within a strict window of a given reference lap. 3. complete a high-speed overtaking manoeuvre in a controlled way.
Of the 11 teams on the grid, only seven managed to pass all the tests and earn a place in Grand Final qualifying – Rapson among them, lining up alongside some of the most experienced outfits in the field for three qualifying runs.
During the second qualifying run, however, Rapson’s car suffered a crash. As a result, the team was unable to record a valid lap within 115% of the reference time (the fastest lap of the fastest team), which meant they did not advance to the multi-car Grand Final. Instead, Rapson competed in the Silver Race, where the team ultimately finished fourth.
“Naturally, our sights were set on the Grand Final grid – and we showed in qualifying that we can run with the front-runners.
In the second run we took a heavy hit, but we regrouped and still came away with a solid race result and a genuine innovation milestone. For a first A2RL race weekend, finishing and learning was more important than a hero lap – this is the solid foundation we can now build on.”
concludes Ármin Bogár-Németh, CEO of Rapson.
Partners behind Rapson
Rapson’s A2RL programme is supported by a focused group of partners who help turn ideas from simulation into real laps on track – from deep engineering expertise, through the infrastructure needed to train its AI driver, all the way to cybersecurity.
The Bosch Group - global engineering expertise, presence in Hungary
The Bosch Group is a leading global supplier of technology and services, one of the most innovative companies in the industry. Its operations are divided into four business sectors: Mobility, Industrial Technology, Consumer Goods, and Energy and Building Technology. Bosch uses its proven expertise in sensor technology, software, and services to offer customers cross-domain solutions from a single source. In Hungary, the Engineering Center Budapest has a constantly-rising profile in Bosch’s world-scale developments and plays an important role in the development of automated and electrified mobility. It is also one of the foremost research, development and test facilities for automotive electronics in the Bosch Group.
GanzKK – the Budapest engine behind Rapson’s AI development
Ganz Kapcsoló- és Készülékgyártó Kft. (GanzKK) is a Hungary-based company specialising in low-voltage electrical equipment and systems integration, and is also active in the fields of electromobility and rail vehicle manufacturing. Among other forms of support, GanzKK provides Rapson with high-performance compute capacity, which forms the backbone of large-scale simulation, data processing and AI training. As a result, a significant part of the “brain” behind Rapson’s AI driver is designed, built and operated in Budapest.
Riptides – protecting machine identities in Rapson’s AI stack
Riptides is a cybersecurity partner that focuses on “non-human identities” – the machine-to-machine connections between services and workloads in modern AI systems.
Together, Bosch Group, GanzKK and Riptides form an ecosystem that enables Rapson to put its AI driver on the starting grid of the world’s largest autonomous racing series.
About Rapson
Rapson is a Budapest-based deep-tech racing team and startup, born out of years of engineering in top-level motorsport, automotive and autonomous systems. The team’s mission is to develop superhuman motion intelligence – AI that can make fast, safe, optimal decisions in the physical world – starting in extreme environments like autonomous racing and then bringing that capability into everyday traffic and road vehicles.
Rapson is a competitor in A2RL, racing an autonomous Super Formula–spec car at Yas Marina Circuit in Abu Dhabi and contributing to the global push to bring advanced autonomy from simulation into the real world.
BACKGROUND INFO
Global grid of autonomous innovators
The 2025 A2RL car race brought together 11 teams spanning top universities, national research institutes, industry consortia and deep-tech startups – all racing the same EAV Super Formula platform, but each with its own AI “brain” and philosophy of autonomy.
  • TUM – Germany
  • PoliMOVE – Italy
  • Unimore Racing – Italy
  • Constructor – Germany
  • Kinetiz – United Arab Emirates / Singapore
  • Code 19 – United States
  • Fly Eagle – China / United Arab Emirates
  • TII Racing – United Arab Emirates
  • FR4IAV – France
  • TGM GP – Japan
  • Rapson – Hungary
Race format: time trials, Silver Race and six-car Grand Final
A2RL’s race is designed to test both raw pace and system robustness.
  • Grand Final qualifying
    Before the time trials, every team had to clear a series of quality gates designed to prove safety and core capabilities. Among these, the reference lap time and the high-speed overtake manoeuvre were the most demanding for the field. Ultimately, only seven teams – including Rapson – succeeded in passing all gates and became eligible for Grand Final qualifying.
    After this, the event moved into Grand Final qualifying time trials. In these sessions, each team has a limited number of “runs” to set a lap time. This is where Rapson’s qualifying crash occurred: the car went off during the second run, which meant the team could not record a lap within the tight 115% pace window of the reference time that is required to reach the multi-car Grand Final.
  • Silver Race – time-trial showdown
    Teams that do not meet the Grand Final threshold – including Rapson after its qualifying crash – compete in the Silver Race. This is a time-trial format: one car on track at a time, with the AI driver pushing against the clock to set the best possible laps, while engineers focus on safe operation and data gathering.
  • Grand Final – autonomous race start to finish
    The highlight of the event is the six-car Grand Final, where the top qualifiers line up together for a multi-lap autonomous race. Here, AI systems must handle wheel-to-wheel racing, traffic, and strategy – not only driving fast laps, but also managing overtakes, defensive lines and changing conditions without any human driver in the cockpit.
This format allows A2RL to showcase the full spectrum of autonomous performance: from single-car precision in time trials to full racecraft in the Grand Final.
The A2RL race car: Super Formula performance, autonomous brain
The A2RL car that Rapson races is a Dallara-built Super Formula SF23–derived chassis, adapted specifically for autonomous competition. It is one of the fastest single-seaters outside Formula 1, capable of approaching 280 km/h on the Yas Marina Circuit.
Under the bodywork, the car combines proven motorsport hardware with an AI “brain”:
  • Chassis & powertrain
  • Modified Super Formula SF23–based monocoque (A2RL EAV platform)
  • Turbocharged 2.0-litre four-cylinder engine based on Honda’s K20C1, driving the rear wheels through a six-speed gearbox
  • High-downforce aero package and Yokohama Advan slick tires for peak cornering performance
  • Sensors: how the AI “sees” the world
  • 7 cameras providing near-360° visual coverage around the car
  • 4 radar units to measure relative speed and distance, especially in poor visibility
  • 3 lidar units that scan the environment with laser pulses to build a precise 3D view of the track and other cars
  • GNSS (Global Navigation Satellite System), IMU (Inertial Measurement Unit) and optical ground-speed sensor to help estimate the car’s exact position, speed and orientation on the circuit
  • Actuators & control: how the AI “drives”
  • Fully drive-by-wire steering and braking, using electro-hydraulic actuators to execute the AI’s commands with millisecond precision
  • Brake-by-wire system with Brembo carbon brakes to handle the extreme deceleration typical of top-level single-seaters
  • Compute & connectivity
  • A high-performance on-board computer stack that runs the teams’ autonomous driving software in real time
  • High-bandwidth telemetry link between car and pit wall, enabling engineers to monitor sensor data, AI decisions and vehicle health live during each run
Within this framework, all teams receive the same vehicle, vehicle setup, hardware platform and low-level software, and the real competition is in the software: perception, planning, control and race strategy algorithms that can understand the environment, handle changing grip and traffic, and still deliver fast lap times.
Media Resources and Contact
Download Press Materials
Access the complete press release and high-resolution imagery.
All materials are available for editorial use.
Stay Connected
Follow Rapson's journey in autonomous racing and stay updated on the latest developments in AI-powered motorsport technology.
  • Emergency media line: +36 30 080 4633
Social Channels:
"Racing has always been about pushing the limits of what's possible. With autonomous systems, we're not just competing—we're defining the future of intelligent mobility."
— Rapson Racing Team
© 2025 Rapson. All rights reserved. This press release and associated materials are provided for editorial use only. Rapson, the Rapson logo, and related marks are trademarks of Rapson Kft. A2RL and Abu Dhabi Autonomous Racing League are trademarks of ASPIRE.