Swiss-based scientists are developing computer vision algorithms to help drones 'see' and understand their environment. Matthew Stock reports.
Drones working together to build a 3D model of two statues in real-time, using computer vision technology. Researchers in Zurich are teaching unmanned aerial vehicles to 'see' to help them better understand their surroundings. Their low-cost, off-the-shelf, drones are equipped with a single camera and inertial sensor. But it's their smart algorithms that give them the power to perform more complicated tasks without a human at the controls. SOUNDBITE (English) PROFESSOR MARGARITA CHLI, LEADER OF THE VISION ROBOTICS LAB AT ETH ZURICH, SAYING: "We start off with zero knowledge about where the drone is. The pilot clicks a button so that the drone takes off, as soon as the autonomous mode is on the drone starts perceiving its environment, it starts building a map of its visual landmarks on the goal.... and judging how these move - how these landmarks move - from one image to the next. Then the drone is really reasoning about its motion; essentially the motion of the camera." The map it builds is the basis for the drone's understanding of space. It can now plan a collision-free path and complete missions autonomously. Researchers hope drones working in teams could one day be used for missions including search-and-rescue, industrial inspection and crop monitoring. Or, like with these statues, digitising an archaeological site. SOUNDBITE (English) PROFESSOR MARGARITA CHLI, LEADER OF THE VISION ROBOTICS LAB AT ETH ZURICH, SAYING: "Drones can really offer a unique platform that moves very agile and very fast in the environment. It really moves in 3D so it can really fly up and quickly gather the overview of, for example, a disaster area." With demand for drones booming, the race is on to on to develop ever-more intelligent and autonomous systems. UAVs have recently been trialled to spot poachers in Africa, deliver medical assistance to remote areas, and even to deliver pizzas.