Drones to help rescue people lost in the wild

Geneva, Feb. 11: Scientists have developed an artificial intelligence software for drones to autonomously recognise and navigate through complex environments, and help quickly rescue people lost in forests and mountain areas.

The advance means drones could soon be used in parallel with rescue teams to accelerate the search for people lost in the wild, researchers said.

Every year, thousands of people lose their way in forests and mountain areas. Drones can effectively complement the work of rescue services teams, researchers said.

Since they are inexpensive and can be rapidly deployed in large numbers, they substantially reduce the response time and the risk of injury to missing persons and rescue teams alike.

Researchers, including those from the Dalle Molle Institute for Artificial Intelligence and the University of Zurich developed a software that allows drones to autonomously detect and follow forest paths.

“While drones flying at high altitudes are already being used commercially, drones cannot yet fly autonomously in complex environments, such as dense forests,” said Davide Scaramuzza from the University of Zurich. “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,” Scaramuzza said.

The drone observes the environment through a pair of small cameras, similar to those used in smartphones.

“Instead of relying on sophisticated sensors, their drone uses very powerful artificial- intelligence algorithms to interpret the images to recognise man- made trails,” said Alessandro Giusti from the Dalle Molle Institute for Artificial Intelligence in Switzerland. If a trail is visible, the software steers the drone in the corresponding direction.

“Interpreting an image taken in a complex environment such as a forest is incredibly difficult for a computer,” said Giusti.

Researchers used a Deep Neural Network, a computer algorithm that learns to solve complex tasks from a set of “training examples,” much like a brain learns from experience. In order to gather data to “train” their algorithms, the team hiked several hours along different trails in the Swiss Alps and took more than 20 thousand images of trails using cameras attached to a helmet.

When tested on a new, previously unseen trail, the deep neural network was able to find the correct direction in 85 per cent of cases; in comparison, humans faced with the same task guessed correctly 82 per cent of the time.