Perching Robot Glider Nails Simulated Powerline Landings

This drone is learning how to make pinpoint perching landings

2 min read
Perching Robot Glider Nails Simulated Powerline Landings

The biggest thing holding drones back right now (especially small, inexpensive drones) is arguably battery life. It's really bad. For a drone that can hover and carry any sort of payload, you're looking at 10, maybe 20 minutes tops. And even fixed-wing drones don't do all that much better. This is such an issue that CyPhy Works has developed drones that are continuously supplied with power through a tether, and there are other, slightly crazier (or less immediately practical, let's say) ideas about how to provide drones with power on the fly, like lasers or midair wireless power transfer

One less crazy idea is to just have drones perch: that is, to spend as much time as possible not flying by finding somewhere near where they need to be that they can land and sit. And wouldn't it be great if drones could recharge themselves by perching on powerlines and harnessing the magnetic fields that they emit?

MIT has been working on drones that can perch on powerlines for a few years now. Perching isn't so hard for an aircraft that can hover: we've seen quadrotors do it already. But for fixed-wing drones, perching becomes much more complicated, since the drone needs to go from forward flight to something very close to zero velocity in a very small window of space and time.

It's a difficult problem, but certainly not impossible, because we see birds do it over and over again every day, by executing a stall maneuver right before they reach the location that they want to perch on. When talking about aircraft, a stall simply means that the angle of attack of a wing gets high enough, coupled with an airspeed that's low enough, that the air moving over the top of wing separates from the wing itself. This results in an abrupt decrease in lift and increase in drag, and usually, it's a very bad thing, unless you can pull it off on purpose at exactly the right moment, like just in front of a powerline, as a glider from MIT can do:

In these experiments, the location of the line is known to the system, but the glider itself  has on-board sensors and electronics that can plan and execute the perching maneuver in real-time. The sensors are not especially complex, and the actuation on the aircraft is minimal: none of the complex wing morphology that birds have is necessary to make the perching maneuver work. What's also notable is that the perching is successful (95 percent of the time) even in variable starting conditions, as you'd get when you've got a human doing the launching. This is important, because outside of the lab, the aircraft is going to have to deal with a lot of additional uncertainties, like wind.

While it's tempting to focus on the powerline harvesting aspect of this research (like, um, we did), the researchers are emphasizing that what they've really shown is the simplicity of the perching maneuver itself. In other words, a lot of what birds have going on with their complex wings and feathers and build-in neural control system may not be necessary for a maneuver that seems, at first glance, like it should be quite difficult. This means that potentially, all kinds of UAVs, no matter what their level of complexity, might be able to substantially improve their capabilities and versatility by adding perching to their repertoire of tricks.

[ Paper ] via [ MIT ]

The Conversation (0)

How the U.S. Army Is Turning Robots Into Team Players

Engineers battle the limits of deep learning for battlefield bots

11 min read
Robot with threads near a fallen branch

RoMan, the Army Research Laboratory's robotic manipulator, considers the best way to grasp and move a tree branch at the Adelphi Laboratory Center, in Maryland.

Evan Ackerman
LightGreen

This article is part of our special report on AI, “The Great AI Reckoning.

"I should probably not be standing this close," I think to myself, as the robot slowly approaches a large tree branch on the floor in front of me. It's not the size of the branch that makes me nervous—it's that the robot is operating autonomously, and that while I know what it's supposed to do, I'm not entirely sure what it will do. If everything works the way the roboticists at the U.S. Army Research Laboratory (ARL) in Adelphi, Md., expect, the robot will identify the branch, grasp it, and drag it out of the way. These folks know what they're doing, but I've spent enough time around robots that I take a small step backwards anyway.

The robot, named RoMan, for Robotic Manipulator, is about the size of a large lawn mower, with a tracked base that helps it handle most kinds of terrain. At the front, it has a squat torso equipped with cameras and depth sensors, as well as a pair of arms that were harvested from a prototype disaster-response robot originally developed at NASA's Jet Propulsion Laboratory for a DARPA robotics competition. RoMan's job today is roadway clearing, a multistep task that ARL wants the robot to complete as autonomously as possible. Instead of instructing the robot to grasp specific objects in specific ways and move them to specific places, the operators tell RoMan to "go clear a path." It's then up to the robot to make all the decisions necessary to achieve that objective.

Keep Reading ↓ Show less