Drones are becoming an amazingly active and common component of our lives. Be it the military that uses them for reconnaissance missions or for engaging the enemy or be it the civilians who make use of drones for either entertainment or for accessing areas that are otherwise hard to get to. However, drones are unable to negotiate tight spaces with unpredictable movements and rely on the operator. This renders them pretty much useless without a dedicated operator who is managing the flying and even then, the speeds that can be achieved are quite low.
This is where DARPA comes in with the Fast Lightweight Autonomy (FLA) program. As part of this program, a fully loaded quadcopter was recently flown indoors through an obstacle course at a speed of 72km/h. The test flight was carried out to demonstrate the autonomous control along and shall help in development of UAVs that are capable of navigating cramped spaces without relying on GPS or external guide.
Mark Micire, DARPA program manager said, “Very lightweight UAVs exist today that are agile and can fly faster than 20 meters per second (45 mph), but they can’t carry the sensors and computation to fly autonomously in cluttered environments. And large UAVs exist that can fly high and fast with heavy computing payloads and sensors on board.” According to him, the finding of the perfect mix of small size, weight and power for an air vehicle that features limited onboard computing power while being able to carry out a complex mission autonomously is what makes the FLA program so challenging and adrenaline-inducing.
DARPA’s FLA program has its eyes set on the creation of a demonstrator UAV system that features a size that allows user to put the UAV through the building window and is still capable of navigating, autonomously, the interior while maintaining a speed of 45mph without any help from GPS or operator. According to DARPA, the salvation lies in a new class of algorithms that consume less processing power and feature reduced dependency on a human operator while still being able to tackle rooms, stairs, obstacles and corridors.
The aim of the program is to allow UAVs to be able to carry out work in spaces that offer poor navigational signals and are cluttered. This would allow the operator to focus on higher level tasks rather than being occupied with navigation of drone. The test flights were carried out in an air hangar at Otis Air National Guard Base, Cape Cod, Massachusetts. The air hangar was converted into a maze by making use of simulated wall and boxes. Three research teams made use of a commercial DJI Flamewheel 450 airframe that was equipped with E600 motors along with 12-in propellers and a 3DR Pixhawk autopilot. Apart from all that, it also sported high-definition onboard cameras, LIDAR, inertial measurement units, sonar and a variety of other sensors.
The first series of tests was carried out at low and high speeds and the operator acted merely as an observer. According to DARPA, now that the initial data gathering has been completed, the obstacle course will be given an upgrade to make it more realistic and challenging for the drone. DARPA plans on incorporating this technology to ground and marine vehicles as well for environments where GPS reliability becomes shady.
Micire further added, “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 vehicle’s’ mass to make extremely tight turns and abrupt maneuvers at high speeds.”