When I was eight, I got lost in the woods on a camping trip with my mom. Rangers eventually found me and hiked me out, which is actually kind of miraculous. I was a couple miles from where they assumed I'd be, trudging haplessly through a remote backwoods area on unmarked terrain. A proverbial redheaded needle in a haystack.
Every year hundreds of thousand people get lost in the wild worldwide. Some are injured, some have just lost their way. Drones, which are already being used for search and rescue during natural disasters, could make an alluring complement to human rescuers in national parks and wilderness areas. Drones are relatively cheap and could be deployed in large numbers very quickly, reducing response times in situations where every minute counts.
But flying drones between trees is risky. It says so right on the box: Avoid trees or you will crash your quadcopter. I've done it. One low hanging leaf is sometimes enough to send your precisely balanced airframe toppling to the ground. There's just no margin for error.
Which is why flying in forests is going to take some absolutely astounding artificial intelligence and path planning.
A group of Swiss researchers from the Dalle Molle Institute for Artificial Intelligence, the University of Zurich, and NCCR Robotics has developed Artificial Intelligence software to teach a small quadrocopter to recognize and follow forest trails all by itself, staying low enough to avoid tree canopies.
This research could soon be used in parallel with rescue teams to search for people lost in the wild faster than would be achievable by human rescuers alone. Of course a drone that can follow trails in dense jungles could also have more controversial military and intelligence applications, as well, so nice bit of PR awareness by the Swiss.
The drone used by the researchers observes the environment through a pair of small cameras, similar to those in your smartphone. Instead of relying on sophisticated sensors, their drone uses very powerful artificial-intelligence algorithms to interpret the images to recognise man-made trails. 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; sometimes even humans struggle to find out where the trail is," says Dr. Alessandro Giusti from the Dalle Molle Institute for Artificial Intelligence--presumably trying to make me feel better about my eight-year-old self.
The Swiss team solved the problem using a so-called 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 enough data to "train" their algorithms, the team hiked several hours along different trails in the Swiss Alps and took more than 20,000 images of trails using cameras attached to a helmet. Pretty cool gig, right?
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% of cases; in comparison, humans faced with the same task guessed correctly 82% of the time.
The research team warns that much work is still needed before a fully autonomous fleet will be able to swarm forests in search of missing people, but that day is getting closer.
Prof. Luca Maria Gambardella, director of the Dalle Molle Institute for Artificial Intelligence in Lugano, explains: "Many technological issues must be overcome before the most ambitious applications can become a reality. But small flying robots are incredibly versatile, and the field is advancing at an unseen pace. One day robots will work side by side with human rescuers to make our lives safer: this is a small but important step in that direction."