Drone flight algorithm beats all human pilots in test race
Image credit: University of Zurich
An algorithm for automatically piloting drones has been developed that can outperform human pilots for the first time.
Due to their limited battery, speed is of the essence when drones are completing tasks like searching for survivors on a disaster site, inspecting a building or delivering cargo. The routes they take are sometimes complex and narrow, requiring precision flying.
Up to now, the best human drone pilots would always outperform autonomous systems in drone racing, according to researchers at the University of Zurich (UZH) in Switzerland.
But they have now created an algorithm that can find the quickest trajectory to guide a quadrotor – specifically a drone with four propellers – through a series of waypoints on a circuit.
“Our drone beat the fastest lap of two world-class human pilots on an experimental race track”, said researcher Davide Scaramuzza.
“The novelty of the algorithm is that it is the first to generate time-optimal trajectories that fully consider the drones’ limitations.”
Previous works were not optimal because they relied on simplifications of either the quadrotor system or the description of the flight path.
Philipp Foehn, PhD student and first author of the paper, said: “The key idea is, rather than assigning sections of the flight path to specific waypoints, that our algorithm just tells the drone to pass through all waypoints, but not how or when to do that”.
The researchers had the algorithm and two human pilots fly the same quadrotor through a race circuit.
They employed external cameras to precisely capture the motion of the drones and - in the case of the autonomous drone - to give real-time information to the algorithm on where the drone was at any moment. To ensure a fair comparison, the human pilots were given the opportunity to train on the circuit before the race.
The algorithm-driven drones won consistently, beating humans across all of its laps, and the performance was more consistent. Once the algorithm found the best trajectory it was able to reproduce it faithfully many times, unlike human pilots.
The researchers said they needed to make it less computationally demanding before it can be applied to commercial applications.
Currently, it takes up to an hour for the computer to calculate the time-optimal trajectory for the drone. Also, at the moment, the drone relies on external cameras to compute where it was at any moment. In future work, the scientists want to use onboard cameras.
The UK began testing a drone delivery network in March that is intended to facilitate fast delivery of Covid-19 supplies and blood samples.
Sign up to the E&T News e-mail to get great stories like this delivered to your inbox every day.