AirBurr MAV Navigates by Bouncing Off Walls and Floors

When you have a robot that can't crash, navigation gets a whole lot easier

2 min read
AirBurr MAV Navigates by Bouncing Off Walls and Floors

A lot of UAV research is focused on making flying robots that can navigate by themselves using sophisticated sensor systems, intelligently avoiding crashing into things. This is a fantastic goal to have, but it's not easy. EPFL is doing away with just about all of that with a new version of AirBurr, a robot that's specifically designed to run into everything and crash all the time, building maps as it does so.

As we saw last June, AirBurr has undergone a remarkable evolution since 2009. And even in 2012, they were only on version 8, while the current version is up to 11. AirBurr is a coaxial UAV that is totally comfortable with collisions, thanks to its shock absorbing roll cage and self-righting mechanism:

Its rigid central core is surrounded by specially-designed tetrahedral-shaped springs that buckle to efficiently absorb impact energy. The springs protect the AirBurr from impacts with obstacles and can be used to physically interact with objects while in flight. If a collision results in a fall to the ground, the robot's Active Recovery System, comprised of a system of spring-loaded carbon fibre legs, allow it to return to an upright position and take off again.

Here it is in action:

Obviously, having just four sensors makes AirBurr kinda terrible at obstacle avoidance, but the simple fact is that it just doesn't matter: it may not be efficient at finding its way down a hallway, but it does so with an absolute bare minimum of sensors, and it wouldn't care if the hallway was pitch black or full of smoke or otherwise a place in which conventional vision would be out of luck. This vastly increases the number of environments in which AirBurr can be used.

The mapping behavior is especially cool, and if the resulting light paintings remind you of anything, it's because AirBurr employs a random direction algorithm that's similar to the one used by some robotic vacuum cleaners.

This sort of behavior is based in no small part on insects, which also have very primitive sensing systems combined with body structures that allow them to survive numerous collisions. Bugs may not be particularly smart, but as it turns out, big brains and complex sensors aren't always necessary for robust flight and navigation.

We'll see more of this research at ICRA in May from the EPFL team (which includes Briod Adrien, Adam Klaptocz, Kornatowski Przemyslaw Mariusz, and Zufferey Jean-Christophe), but there's a hint on EPFL's website as to where the researchers are taking this: they'll be presenting a paper entitled "A Perching Mechanism for Flying Robots Using a Fibre-Based Adhesive." Cool!

[ AirBurr ]

Thanks Adam!

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

“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.

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

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