18-04-2025
Watch: AI drone beats human pilot in Abu Dhabi's $1 million prize pool autonomous race
Abu Dhabi: AI is no longer learning from humans – it's starting to beat them. In a major breakthrough for autonomous flight and aerial robotics, an AI-powered drone has outpaced human pilots in a global competition held in Abu Dhabi.
The Abu Dhabi Autonomous Racing League (A2RL), a project under the Advanced Technology Research Council (ATRC), in collaboration with the Drone Champions League (DCL), hosted one of the world's most sophisticated drone races at ADNEC Marina Hall.
A total of 14 international teams made it to the finals week, competing for a $1 million prize pool. Teams from the UAE, Netherlands, Austria, South Korea, the Czech Republic, Mexico, Turkey, China, Spain, Canada, and the US represented university labs, research institutes, and deep-tech startups.
The highlight? MavLab, from the Delft University of Technology in the Netherlands, secured victories in three out of four competitions. They clinched the AI Grand Challenge with their drone completing two laps of the 170-metre course in just 17 seconds. MavLab won the world's first AI-only drag race, demonstrating straight-line speed and precision under intense acceleration.
In a landmark moment, MavLab's autonomous drone defeated three top DCL champion pilots in a head-to-head AI-versus-human showdown. With precision flying, the AI-powered drone edged out its human-piloted rivals in thrilling contests. (Watch the video)
'Winning three top titles is a huge milestone for our team,' said Christophe De Wagter, team principal of MavLab. 'I always wondered when AI would be able to compete with human drone racing pilots in real competitions. I'm extremely proud of the team that we were able to make it happen already this year.'
The results, he underlined, validates years of research and experimentation in autonomous flight.
'To see our algorithms outperform in such a high-pressure environment and take home the largest share of the prize pool, is incredibly rewarding,' De Wagter noted.
Meanwhile, Technology Innovation Institute (TII), Abu Dhabi, bagged the multi- autonomous drone AI race in a high-speed challenge that tested coordination, navigation, and collision avoidance between multiple autonomous units.
How did they race?
Each team raced a standardised drone equipped with a compact computing module, a forward-facing camera and an inertial measurement unit. With zero human input, the drones relied solely on real-time processing and AI-driven decision-making, hitting speeds of more than 150 km/h through a challenging course. The race environment pushed the boundaries of perception-based autonomy, with wide gate spacing, irregular lighting, and minimal visual markers. To raise the difficulty, the event used rolling shutter cameras – further testing each team's ability to achieve fast, stable performance in visually sparse conditions. This was the first time an autonomous drone race of this scale and complexity was held under such constraints, underscoring the technical sophistication of the event.
