The July 2022 issue of IEEE Spectrum is here!

Close bar

Video Friday: AR Drone Stunt, Real Transformer, and Futurama Justice

What better way to kick off the weekend than with a flaming flying robot?

3 min read
Video Friday: AR Drone Stunt, Real Transformer, and Futurama Justice

This week's Video Friday is kicking off with a long list of things not to do with an AR Drone unless you have the budget of a successful French wireless device manufacturer named after a bird.

Parrot's AR Drone is, as far as robots go, nearly indestructible. To highlight this fact, Parrot spent more than four months preparing and filming a super slow-mo video of a drone flying through fire, water, and dust. It's epic, in the way that only things with a big budget can be epic:

Some things to note about this video: there's no CGI, it's all real drones flying through real fire and other real stuff that robots generally don't get along with. That's a real crash into a brick wall at the end. And remarkably, none of the 25 drones used during the filming suffered any major injuries. Trust me, I've been playing with one of these things for a little while now, and they can take a beating. You can see more in this behind-the-scenes video:

For just $300, this is an incredible robot. And especially now that you can control it with ROS, it's worth a serious look.

[ AR Drone ]

Teams of robots that work underground could learn a thing or two from ants. Insects are experts at working cooperatively in small, dark spaces, and Georgia Tech researchers are trying to figure how they move around:

[ Georgia Tech ]

This ad for the Chevy Volt came out a couple months ago, but I missed it. Probably because it's an ad. And I don't watch ads. But this ad features a cute robotic dog, and I'm also a big fan of the Volt, so it's worth a look:

You know, every time I see CGI footage of robots like this, I can't help but think something along the lines of, "holy moley that robot is going to be utterly impossible for a very, very long time." Sigh.

Via [ AdWeek ]

We met UC Berkeley's STAR "sprawl-tuned" robot back in February, but here's an updated video showing what the little bot was primarily intended to be used for: playing pool.

[ Biomimetic Millisystems Lab ]

Hey look, it's a little robot car. But just wait until 40 seconds in:

WHOA. I had to watch that first transforming move like eight times in a row to convince myself that it wasn't a.) fake or b.) sped up, but I think it might be legit. This is a custom transformer robot by Kenji Ishida, who wants to have one of these THAT YOU CAN RIDE IN by 2030.

[ Brave Robotics ]

Humans are awful at languages. We've got thousands of them (close to 7000, in fact), and even if you speak the same language as someone else, regional accents and dialects may make them virtually incomprehensible. This (among other reasons) is why speech recognition is so difficult for robots, and the ROILA project is solving this problem by inventing a brand new language exclusively for humans interacting with robots and software—a kind of Esperanto for human-machine communication:

There's about 800 words in total, each one constructed by a genetic algorithm to be as different from the others as possible so that robots can understand them without any trouble. Check out the language guide here.

[ ROILA ]

Well, this is seriously awesome: turns out you can hacksaw a Sphero in half and give it some eyeballs:

Probably not the best thing to try unless you know exactly what you're doing, though.

[ RoseRolls ]

Curiosity has been busy on Mars. How busy? This busy:

[ MSL ]

This video of a group of dancing NAOs from Aldebaran and MIT is unique because it's demonstrating a robust synchronization program that allows individual robots to drop out and join back in without missing a beat:

Read the paper here.

Remember the Costa Concordia? Team BlackSheep takes us on a quadrotor tour of the still partially sunken cruise liner:

[ Team BlackSheep ]

Last month, Futurama co-producer Patric Verrone gave an absolutely fantastic talk at the We Robot conference at Stanford. A video of the talk is now online, and that's the good news. The bad news is that none of the clips of Futurama itself made it into the YouTube video (for reasons you can probably imagine). However, hardcore Futurama fans should be able to picture all of the scenes that Patric talks about, and even if you're not as familiar with the show as you should be, it's still an awesome talk about the ethics of robots and justice in a cartoon future a thousand years from now.

Via [ RoboHub ]

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

“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