The July 2022 issue of IEEE Spectrum is here!

Close bar

The invasion ofsocial robotscontinues. This week, Japanese robot maker Vstone and telecom company NTT announced plans to market little humanoids that can interact with people and also with devices around the house. In a press conference in Tokyo, the desktop-size robots, called CommU and Sota, held conversations with two lifelike female androids and, later, a male android. (Hmm, this last one, it’s possible that it was a real man, but we’re not sure.)

The little talking robots look really cool, especially the way the turn their heads and eyes in the direction of the person or robot that’s speaking. But this paragraph in the Japan Times article left us a bit confused:

CommU and Sota cannot understand what people are saying and are just acting out programmed conversations. They can only tell whether an individual is speaking or not.

We hope a future upgrade will let the robots understand what you’re talking about, so you can have an actual, you know, conversation with them.

[ Nikkei ] and [ Japan Times ]

Thanks Glenn!

We have no additional information on this video, besides that it appears to show a human-sized version of PIBOT using a full scale flight simulator:

It’s obviously not using vision, but based on output from the simulator, it can convincingly take off and land.

[ USRG ]

If you’re looking forward to getting a Jibo and then immediately trying to do your own stuff with it, here’s what you have to look forward to:

[ Jibo ]

McGill’s robotics program looks cool enough that we can forgive them all for being Canadians (and therefore funnier than the rest of us):

[ McGill University ]

Hinamitetu now has a robot that can do vault:

[ YouTube ]

This was somehow not a thing that we knew existed: Robocon, a competition for autonomous badminton robots. YouTube has some robot-on-robot matches, but the more interesting video is this exhibition between a pair of robots from UESTC and Dong Jiong, who won a silver medal in badminton in Atlanta:

Not bad, right?


This video is from 2012, and it doesn’t show any amazing new research or anything. But sometimes, little fluffy kitties are just, like, necessary, you know?

[ Hexbug ]

HERB, of Carnegie Mellon’s Personal Robotics Lab, and CURI, of Georgia Tech’s Socially Intelligent Machines Lab, learn trajectory constraints from demonstrations.

How not to spill stuff, in other words:

[ HERB ]

It's Back to the Future at RoboSub 2015:

[ RoboSub 2015 ]

Researchers from MIT, CMU, and ABB presented a paper on “extrinsic dexterity” for robots (using the environment to assist with grasping) at ICRA 2014. Here’s an update video from MIT that looks like it’s probably intended to be attached to a press release that hasn’t been, um, released:

[ MIT ]

It’s (still) good to see that DRC teams haven’t all called it quits and mothballed the robots that they worked so hard on, and ESCHER is now able to handle stairs with no trouble at all:

[ TREC ]

Thanks Robert!

WPI-CMU's DRC team also sent us a video this week, featuring “less robot abuse” than the one they sent us last week. Not sure how to feel about that, honestly, because well-intentioned and practical robot abuse is almost always fun to watch, and not seeing ATLAS bite it is always a vague disappointment:

[ Team WPI-CMU ]

Thanks Siyuan!

The continually evolving design for UMD’s solar-powered flapping wing Robo Raven (now on version 3, subversion 4) has always struck me as supremely elegant:

[ UMD ]

Is there anything that ATRIAS can’t do? Besides maybe stand still?

[ OSU ]

SenseFly felt the need to test its eXom drone at high altitude on a mountain in Switzerland:

Next, we’re expecting more testing to see how well eXom performs on tropical beaches.

[ eXom ]


Via [ Gizmodo ]

TIAGo, the mobile manipulator: it is a service robot created by PAL Robotics to provide the best help in universities and research institutions. TIAGo is an ideal platform for research, as it combines navigation, perception, manipulation and human-interaction abilities. Watch TIAGo's prototype in action on this video.

It’s great to see the physical version of TIAGo, since we’ve mostly just seen still images and renderings up until now. The big question is, will TIAGo be able to out-UBR-1Fetch Robotics?

[ PAL Robotics ]

Throwing a drone and catching it back like it’s a boomerang is something that you should absolutely not try with any drone except for Gimball:

[ Flyability ]

Pay attention, Kuka, because this is a ping pong robot that really works:

[ Omron ] via [ BB ]

We’re continually hoping for great things from Anki. We haven’t seen ’em yet, but we’re still hoping.

[ Anki ]

We have no idea who Bart van Overbeeke is, but he’s put together some highlight movies of Tech United Eindhoven at RoboCup 2015:

[ Tech United Eindhoven ]

ETH Zurich’s Autonomous Systems Lab has been developing a solar powered, long endurance UAV that managed to stay aloft for 81 hours straight. To end the week, here’s a video of the test run in REAL TIME!

Kidding, kidding, it’s a 4-minute summary, and there is actually some drama at the end, so it’s worth watching:

[ Atlantik Solar ] via [ DIY Drones ]

Seriously, though, to end the week, here’s a talk from Dr. Luis Sentis and Donghyun Kim from UT Austin’s Human Centered Robotics Lab given just a few days ago at Dynamic Walking 2015:

[ Dynamic Walking 2015 ]
[ UT Austin HCRL ]

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