According to the agency, the Fast Lightweight Autonomy program is focusing on “creating a new class of algorithms to enable small, unmanned aerial vehicles to quickly navigate a labyrinth of rooms, stairways, and corridors, or other obstacle-filled environments without a remote pilot.”
These algorithms would apply to already existing drones which are capable of flying through an open window at 45 miles per hour.
"Birds of prey and flying insects exhibit the kids of capabilities we want for small unmanned aerial vehicles," said Mark Micire, DARPA program manager.
"The goal of the FLA program is to explore non-traditional perception and autonomy methods that would give small UAVs the capacity to perform in a similar way, including an ability to easily navigate tight spaces at high speed and quickly recognize if it had already been in a room before."
Effective bug-size drones could be available in another 15 years.
These machines are tiny. The smallest drone in the world, a robofly, is barely larger than a quarter, and the military is working on one three times smaller.
But while the mechanics are all but worked out, the brains are not. Such small devices can’t carry a large computer, and any machine capable of doing everything that DARPA wants their insect drones to do, has to come with a lot of computing power.
“Since micro rotorcrafts can only carry a few grams of payload including batteries, this has to be accomplished with a very small weight and power budget,” according to a NASA paper titled “Towards Autonomous Navigation of Miniature UAV.”
Autonomous flight, without the aid of operators, GPS directions, or external sensors, will require the drone to process tons of visual data from its onboard camera. And, flying at 45 mph, it will have to process this information quickly.
Essentially, DARPA wants a low-grade artificial intelligence capable of analyzing the world around it.
Stefanie Tompkins, director of DARPA’s Defense Sciences Office, hopes that these new software algorithms could let human operators focus on the larger mission.
“By enabling unmanned systems to learn ‘muscle memory’ and perception for basic tasks like avoiding obstacles, it would relieve overload and stress on human operators so they can focus on supervising the systems and executing the larger mission,” she said.
Once perfected, these drones could surely spot flood survivors and report back to a rescue team on the ground. They could also serve as literal flies-on-the-wall, perching in discreet corners of rooms that they, themselves, have chosen, and recording conversations within.
