Close

Video Friday: PR2 With Nailgun, Snake Bot Tango, and Robot vs Sword Master

Get caught up with non-ICRA, non-DRC videos. We promise!

3 min read
Video Friday: PR2 With Nailgun, Snake Bot Tango, and Robot vs Sword Master
Image: Yaskawa

Over the past few weeks, between ICRA and the DRC Finals, we’ve had enough incredible and exciting and amazing and stupendous robots to last us the rest of the summer. But of course, while we were focused on Seattle and Pomona, other cool robotics stuff was happening. We’re going to get you caught up on some of it today, in a post that we promise will have zero DARPA Robotics Challenge content, because after more than a dozen posts in like five days, we all need just a little bit of a break.

Rather than starting out with a robot and a ninja, check out something that might be even cooler: the latest iteration of the Blind Juggler, created by roboticist and artist Raffaello D’Andrea and his group at ETH Zurich. This machine uses no sensors to toss balls in a circular pattern between a cloverleaf of carefully machined and tilted paddles:

What’s next? A concept that’s yet cooler still:

This is based on real physics, and we’re expecting to eventually see a live implementation.

[ Blind Juggler ]

Now robots and ninjas? NO. Robots and nailguns!

The birdhouse team (Ben Cohen, Mike Phillips and Ellis Ratner) got a PR2 to build birdhouses via a autonomous flexible assembly model.

And nailguns.

[ CMU ]

Okay, now robots and ninjas.

Yaskawa Electric Corporation will continue to transmitting "manufacturing spirit" inherited unbroken in this Corporation to the world at the juncture of its 100th anniversary. "YASKAWA BUSHIDO PROJECT" is the project representing Machii's swordplay exactly the same by using "MOTOMAN-MH24", he retains many world record like "Cutting BB Gun 6mm pellet" and his play is also said God's miracle, while challenging to the performance limits of the industrial robot that integrate the "agility", "accuracy", "flexibility" to high-dimension.

[ Yaskawa Motoman ]

Inside of ninjas, there are squishy goopy things. Inside of robots, there’s all this stuff:

[ KUKA ]

ABB has been working on non-one-big-giant-arm robots that are better at collaborating with humans and not squishing them into goop. The infuriatingly capitalized YuMi uses soft fingers to put apples in boxes, seriously threatening the apple boxing profession:

[ YuMi ]

I hope you’re ready for Thymio to go wireless, because Thymio is going wireless:

€150 gets you one, and upgrade kits are available for existing Thymios.

[ Indiegogo ]

Demos of drones that can follow people invariably take place at ski resorts, on rivers or lakes, in the middle of wide open fields, or other places where there aren’t a whole bunch of things for these flying robots to smash themselves into. It’s nice to see this demo from CyPhy of their LVL 1 drone following someone through the middle of a lot of busyness in Boston:

There’s still time to get one on Kickstarter!

[ Kickstarter ]

We like the tailsitter design for drones because it allows them to transition from hovering (situationally useful but inefficient) to wing-aided forward flight (very efficient for covering distance). The control problem is a more difficult one, but ETH Zurich has cracked it, at least in a motion capture arena:

[ ETH Zurich ]

And now, this:

[ CMU Biorobotics ]

Jennifer Darwin won the 2015 ICRA Humanoids Challenge for demonstrating skills like these:

The robot actively balances and compensates for small disturbances in the slope by using PID controllers to control the inclination of the centre-of-mass in both the left-right and front-back directions. Turning and braking are accomplished through ZMP control.

We use coloured markers to indicate the course; the robot is programmed to stay inside the region marked by the blue and pink flags.

Cross-country skiing uses a heavily-modified walking gait inspired by our previous work on ice skating in 2012. By lengthening the stride, lowering the vertical amplitude, and increasing the period time we were able to develop a gait that allowed the robot to move across natural snow.

[ Manitoba ]

ESA is sending a robot to Mars with a big drill in 2016, with the hope of finding some trace of life. We’ll end this long, long week with a not so long (8 minutes) video about the robot and the mission.

[ ExoMars ]

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