Teaching Drones to Travel in Small spaces by Watching the Birds and the Bees

It really interesting to make note of the different approaches that are used to solve problems.  In this case university research seeks to learn how to teach a drone how to travel though a confined space by emulating the flight patterns of birds and bees.

There are different approaches to solve the problem of flying in a confined space and this one uses an approach called “optical flow”.  The concept of optical flow was introduced by the American psychologist James J. Gibson in the 1940s to describe the visual stimulus provided to animals moving through the world.^[3]

Researchers at the University of Maryland are adapting the techniques used by birds and bugs to teach drones how to fly through small holes at high speeds. The drone requires only a few sensing shots to define the opening and lets a larger drone fly through an irregularly shaped hole with no training.

One commercial product moves in confined space is Flyablility but it uses a different approach.  Another base level technology is found in the DroNet project that talks about traffic patterns and how to emulate them in autonomous vehicles.

The project called GapFlyt, to teach drones using only simple sensors.

Optical flow, creates a 3D model using a very simple camera. By marking locations in each following photo, the drone can tell the shape and depth of holes based on what changed.

As you can see in the video below, the researchers have tried to create a challenging environment in which to test their system. The Bebop 2 drone with an NVIDIA Jetson TX2 GPU moves  around the hole like a bee or a bird and then  through at 2 meters per second.

The development team at the University of Maryland’s Perception and Robotics Group reported that the drone was 85 percent accurate.

It will be interesting to see how this type of base level technology will be applied in commercial and recreational drone usage in the future.