a drone built by NVIDIA researchers navigates the most far-flung, un-mapped places using deep learning and computer vision powered by NVIDIA ‘jetson TX1’ AI supercomputers. initially designed to follow forest trails to rescue lost hikers or spot fallen trees, the low-flying autonomous drone could work in canyons, between skyscrapers, or inside buildings where GPS is restricted, or unavailable.
all images © NDIVIA
to keep costs low, the researchers built their device using an off-the-shelf drone equipped with the NVIDIA ‘jetson TX1‘ and two cameras.
‘our whole idea is to use cameras to understand and navigate the environment,’ says nikolai smolyanskiy, the team’s technical lead. ‘jetson gives us the computing power to do advanced AI onboard the drone, which is a requirement for operating in remote environments.’
the team has already trained it to follow train tracks, and ported the system to a robot-on-wheels to traverse hallways. the drone also avoids obstacles like people, pets or poles. although the technology is still experimental, it could eventually search for survivors in damaged buildings, check stock on store shelves, inspect railroad tracks in tunnels, or be adapted to examine communications cables underwater.
the autonomous drone learned to find its way by watching video that smolyanskiy shot along eight miles of trails in the pacific northwest of america. he captured the video in different lighting conditions using three wide-angle gopro cameras mounted on the left, center and right of a metal bar on a mini segway.
the team now plans to create downloadable software for ‘jetson TX1’ and ‘jetson TX2’ AI supercomputers so designers can build robots that navigate on visual information alone. the idea is to command the robot to travel between two points on any map — whether it’s a google map or a building plan — and have it successfully make the trip, avoiding obstacles along the way.
martin hislop I designboom
jun 16, 2017