Check out the on-demand sessions from the Low-Code/No-Code Summit to learn how to successfully innovate and achieve efficiency by upskilling and scaling citizen developers. Watch now.

The Drone Racing League (DRL) will host a series of races and competitions where autonomous drones will try to beat a professional drone pilot. Teams of university students and other drone enthusiasts will be invited to compete for more than $2 million.

The Artificial Intelligence Robotic Racing (AIRR) Circuit will produce four AI vs. AI races in its inaugural season and will use the same video game-inspired courses as the kinds used by pro drone pilots in the DRL Allianz World Championship, DRL CEO Nicholas Horbaczewski told VentureBeat in an email.

Both the AIRR Circuit and the AlphaPilot Innovation Challenge, which pits human versus machine, will take place during the 2019 season.

The first drone from an autonomous team to beat a human will receive $250,000, and the winning team of the AIRR Circuit will receive $1 million. Full details on team requirements will be made available when applications are accepted in November.


Intelligent Security Summit

Learn the critical role of AI & ML in cybersecurity and industry specific case studies on December 8. Register for your free pass today.

Register Now

“The AIRR Circuit will be the premier, global autonomous drone racing circuit. It will challenge teams of the most talented AI engineers and researchers from around the world to design an AI framework that’s capable of racing a drone — without any pre-programming or human intervention,” Horbaczewski said. “By having AI and humans compete on the same track and comparing their times, we’ll also be able to measure the gap in performance between man and machine, and see how quickly it closes.”

Above: Jet, the 2016 and 2017 DRL Allianz World Champion.

The competition is being funded by Lockheed Martin as part of a multi-year partnership with the DRL, but it’s not the first instance of human-machine competition to be the best drone pilot. Last fall, NASA’s Jet Propulsion Laboratory (JPL) held a race between pro drone racer Ken Loo and a set of autonomous drones that made their decisions based on feedback from cameras. Overall, the drones were slower but more consistent than the human pilot.

Plans to launch an AI-fueled circuit and competition comes just a day before the season premiere of the third season of the Drone Racing League. The premiere will be broadcast at 7 p.m. PT on ESPN.

The 2018 season for human drone pilots will take place in a range of locations, from a biosphere in Arizona and The Adventuredome at Circus Circus Las Vegas to the Allianz Riviera soccer stadium in Nice, France and a BMW factory in Munich, Germany. The DRL season will culminate with a championship competition in Jeddah, Saudi Arabia.

The 2019 season will culminate with a winner-take-all race between the winner of the AIRR Circuit and the DRL world champion.

“In 2019, we’re fairly certain the human pilot will win. By 2020, it’s anyone’s race,” Horbaczewski told VentureBeat in an email, adding that an autonomous drone could join the DRL as soon as 2021. The DRL could also use AI in the future to help human drone pilots improve their skills or for autonomous drone teams to work with simulators.

Above: Standardized drones will be employed by both autonomous systems and humans.

Hundreds of drones will participate in the initial AIRR Circuit on drone courses around the world. Participating teams will be asked to construct their frameworks using Nvidia’s Jetson platform for autonomous vehicles.

From DeepMind’s AlphaGo win to reinforcement learning that teaches autonomous systems to work together like teammates to AI that plays Starcraft or Dota 2, games and challenges have been key to many breakthroughs in artificial intelligence.

VentureBeat's mission is to be a digital town square for technical decision-makers to gain knowledge about transformative enterprise technology and transact. Discover our Briefings.