- The Washington Times - Wednesday, February 10, 2016

It could prove a boon to the thousands of hikers and climbers who become lost or injured on some distant forest trail or remote hillside. Meticulous Swiss researchers have developed a rescue drone which can autonomously recognize and follow passageways through the undergrowth and rough terrain, using sophisticated artificial intelligence. The scientists ultimately envision an entire fleet of the rescue drones “able to swarm forests in search of missing people” - working right alongside their human counterparts.

Find an intriguing high-definition video of the new drone in action here

“While drones flying at high altitudes are already being used commercially, drones cannot yet fly autonomously in complex environments, such as dense forests. In these environments, any little error may result in a crash, and robots need a powerful brain in order to make sense of the complex world around them,” said Davide Scaramuzza, who led the project for the Dalle Molle Institute for Artificial Intelligence and the University of Zurich.

The researchers say that their new breed of quadrocopter could be rapidly deployed in large numbers to complement human rescue teams, and thus reduce response time. The device relies on small cameras and unique software.

“Instead of relying on sophisticated sensors, the drone uses very powerful artificial-intelligence algorithms to interpret the images to recognize man-made trails. If a trail is visible, the software steers the drone in the corresponding direction,” explained fellow researcher Alessandro Giusti.

The Swiss team use a “deep neural network” that prompts a computer to learn to solve complex tasks from a set of training examples, much like a brain learns from experience, they say. Their research was incredibly complex: the team hiked along multiple trails in the Swiss Alps, capturing over 20,000 images of the environment with a helmet-mounted camera.

“The effort paid off: When tested on a new, previously unseen trail, the deep neural network was able to find the correct direction in 85 percent of cases; in comparison, humans faced with the same task guessed correctly 82 percent of the time,” Mr. Giusti reports in his analysis.

The researchers are still working out the technological issues but note that the business of small flying robots is “advancing at an unseen pace.” And their biggest challenge of the moment?

“Now that our drones have learned to recognize and follow forest trails, we must teach them to recognize humans,” says Mr. Scaramuzza.

• Jennifer Harper can be reached at jharper@washingtontimes.com.

Copyright © 2022 The Washington Times, LLC. Click here for reprint permission.

Please read our comment policy before commenting.

Click to Read More and View Comments

Click to Hide