DARPA FLA project drone zooms indoors at 45MPH

DARPA's Fast Lightweight Autonomy (FLA) program has resulted in a small drone that can zip through indoor corridors at just under 45MPH, doing so autonomously and without crashing. A specially crafted small drone was used, and it was equipped with a high-definition camera and a bunch of sensors onboard, enabling it to 'see' and avoid obstacles. The flight testing happened at the Otis Air National Guard Base in Cape Cod.

The FLA program seeks lightweight unmanned UAVs that can hit speeds of 20 meters per second, or about 45MPH. During an initial data collection phase for the program, some quadcopters were tested in a cluttered indoor hanger. The drone was a DJI Flamewheel 450 frame coupled with E600 motors, 12-inch rotors, and the 3DR Pixhawk autopilot.

The testing found the drone was able to hit that 45MPH speed target, as shown in the video above, and also able to perform "initial autonomous" functions like spotting obstacles and avoiding them — sans the help of an operator.

The aircraft hanger where the testing took place had been modified to look like a warehouse, including things like simulated walls and obstacles like boxes. Each time the drone was successful (there were many crashes, DARPA says), the testing warehouse would be upgraded with more clutter and obstacles to make it more complicated. This continued each time the drone successfully passed the test.

The idea is that drones are used already by the military and organizations to, for example, survey stretches of land or assess a situation, such as a fire. Indoor uses are a big goal, one many researchers have been pursuing, but one that is much harder to achieve due to things like tight corners, random clutter and objects, and questionable lighting.

The FLA seeks to solve these issues by developing software and quadcopters suitable for indoor use. During a military mission, natural disaster, or other scenario, the drone could be released through a window and would then guide itself (very quickly) through the structure while others watch a video feed from the camera.

Said DARPA Program Manager Mark Micire:

We're excited that we were able to validate the airspeed goal during this first-flight data collection. The fact that some teams also demonstrated basic autonomous flight ahead of schedule was an added bonus. The challenge for the teams now is to advance the algorithms and onboard computational efficiency to extend the UAVs' perception range and compensate for the vehicles' mass to make extremely tight turns and abrupt maneuvers at high speeds.