Why it matters: Many commercially available drones today can track a cyclist on a trail, but most of them still rely on GPS to keep track of their position. This is a problem in a forest, where a dense canopy can make the signal quite unreliable. A laboratory at Purdue University has now introduced a drone that chases a person through the forest without GPS.
Purdue computer science professor Aniket Bera posted the 44-second video on LinkedIn last Friday. It shows a quadcopter from his IDEAS Lab weaving between trees to keep a person in the frame. And the drone calculates everything on its own computer, without relying on a map of the forest. Bera says it’s important to avoid the cloud because a drone flying hard around a thin branch can’t afford to wait on a server somewhere.
Of course, the person always disappears behind suitcases. To this end, Bera’s post lists a step for occlusion prediction. The drone guesses where its target is moving until it comes back into view.
Navigating without GPS is also exactly the kind of autonomy the US military has poured money into. Last November, Purdue announced that the Army Research Laboratory had signed a five-year, $1.5 million contract with Bera’s lab to build drones and ground robots that can find their way through hostile terrain without satellites. However, it is not clear from the LinkedIn post whether this money was used to finance the forest flights.
In the same post, Bera wrote that his team built every layer of the drone from scratch, from the 3D printed carbon fiber frame to the software on the chip. However, as DroneXL noted, the maps overlaid on the video labeled the drone’s motion planner “SUPER” and its altitude estimate “FAST-LIO.”
These names belong to two open source projects from the MaRS Lab at the University of Hong Kong. SUPER is the laboratory’s own LiDAR-controlled drone. In a Science Robotics paper early last year, the team sent it flying through unknown terrain at more than 20 meters per second, or about 45 miles per hour. A few weeks later, the lab posted the frame’s design files online for free, including designs for 3D printed parts and carbon fiber panels.
The Hong Kong lab is not mentioned in Bera’s post, although SUPER’s code page encourages anyone building on it to cite the team’s papers. There is also no article about the forest demonstration on the IDEAS Lab publications page.
For now, anyone who wants to try building a similar drone can download HKU’s SUPER code as a free download. The lab’s hardware kit also lists all the parts needed, from the Livox Mid-360 LiDAR to the Intel NUC Mini PC that runs the software.
