Portuguese 
English 
Spanish 
2 min of reading 
It Was a Day to Go Down in the History Books. For the First Time, an AI-Controlled Drone Defeated Human Champions in a Racing Competition.
Revealing potential advancements not only for sports but also for other fields such as industrial robotics and vehicle automation.
The study, published in the prestigious journal Nature, documents how the autonomous drone, named Swift, navigated through all obstacles and broke records. And we are talking about a quite complex circuit, 75 meters long with seven gates, where the machines easily reach 100 km/h. If you think Formula 1 is exciting, this drone race would leave you astounded!
Man vs Machine
The humans had an entire week to train. Among the participants was a former world champion in drone racing. But it was the machine that took home the trophy.
-
Home-built airplane by a Minas Gerais engineer uses car parts, such as a power window motor for the flaps and a tachometer on the dashboard. The project took six years, and the aircraft has autonomy to fly up to seven hours non-stop.
-
With 4 engines, an 8,000 km range, torpedoes and anti-ship missiles, the Kawasaki P-1 is the Japanese submarine hunter made to patrol the Pacific and find invisible threats on the seabed.
-
It seems like science fiction, but it already exists: a Swedish electric boat uses a car battery, carbon fiber hull, and smart hydrofoils to fly over the water and travel 105 km in almost total silence.
-
A public school student single-handedly created a machine capable of treating water for up to 50 people using only solar energy. It was awarded third place at one of the most important science fairs in the world.
Inside the Swift
The Swift is not just any drone. It was equipped with a technology called “deep reinforcement learning”, which allowed it to adjust its course in real time and send up to 100 new commands per second. “That’s faster than a blink of an eye,” explains Elia Kaufmann, one of the authors of the study.
While human-controlled drones can benefit from quick adaptation to changes in light and environment, they cannot compete with the processing speed and accuracy of the Swift. However, this speed can be a double-edged sword, as the machine takes significant risks, always operating at the limit.
The success of the Swift is not limited to the world of drone racing. Experts, including military personnel and academics, are keeping an eye on these advancements, as they may impact a much wider range of applications.
Energy Efficiency
According to Kaufmann, the biggest challenge now is how to make these autonomous drones more energy-efficient. After all, what good is speed if you can’t fly for long?
Be the first to react!