NVIDIA Makes Drones Go Wild
Drone flying without GPS, ‘blind’, sounds scar,y dangerous and so 1990s. However, a drone developed by NVIDIA researchers makes its way through unmapped locations using only deep learning and computer vision powered by NVIDIA Jetson TX1 embedded AI supercomputers. Everyone, stand back: the Terminators are coming!
Although initially designed to follow forest trails in an effort to rescue lost hikers or spot fallen trees, the low-flying autonomous drone might actually change the status quo and leave GPS technology to be used by people afraid of Cyrbedyne Systems.
This project manages to point out that there is always an alternative solution to any problem, and I have to admit I was kind of getting sick of GPS and satellites having to track my every move.
The technology is still experimental, but further research might get it looking out for survivors in damaged buildings, inspect railroad tracks in tunnels, and so on.
The team’s already trained it to follow train tracks and ported the system to a robot-on-wheels to traverse hallways. The potential of this technology is incredible, although it seems to have come a bit late; I mean, there aren’t too many wild part of the planet left.
Nonetheless, it is worth mentioning that the device packs the NVIDIA Jetson TX1 and two cameras. With a full AI controlling this flying machine, you wouldn’t have to worry about anything, including battery autonomy.
We’d have random Supercharger-like locations provide the drone with the juice it needs to fulfill its missions. How col would that be?