Video Friday: Your New Buddy, Atlas Fast Walking, and Robot Deception

Don't be deceived. The best robot videos are here, and only here

5 min read
Video Friday: Your New Buddy, Atlas Fast Walking, and Robot Deception
Be my buddy?
Photo: Blue Frog Robotics

After Pepper, Jibo, and Maya, here comes Buddy. Buddy is a mobile, social robot designed as a family companion. It was created by a French company called Blue Frog Robotics, and is now on Indiegogo. We expect to see even more personal robots like these hitting the market in the near future, and that’s a great thing, of course. The biggest question for all of these social robots, though, is whether they’re going to be able to deliver on everything that they’re promising in these promotional videos—the flawless voice recognition and user interaction, the advanced mapping and navigation capabilities, the wide assortment of “apps,” and so forth.

We want these robot buddies to be successful, but there’s a lot riding on this first wave getting it mostly right and not being a disappointment. We’re keeping our fingers crossed.

Like Pepper, Buddy has wheels and can drive around the house. And like Jibo, it’s the size of a coffee maker and has a touchscreen face.

You can pledge for Buddy on Indiegogo for US $650; it’s already 150 percent funded with a month to go.

Indiegogo ]

Yes, Japan has agreed to participate in that giant robot duel we (and everyone else) posted about last week:

We’re jaded on everything all the time (as journalists, it’s our job), but seriously, we’re not exactly sure what people are expecting here. Will these designs even work for hand to hand combat? Like, can they get close enough to each other to land a punch? Or is it just going to be shoving? Our concern here is that it’s going to be yet another one of those things where people get all hyped up and then come to realize that reality is a horrible disappointment, and that that association (robots = disappointment) sticks around beyond this one event. But hey, maybe it’ll be awesome!

We’ve written lots about Aaron Becker’s clever use of robot swarms that obey global commands, although we usually focus on tangible applications. Here’s a fantastic explanation of how robots obeying global commands can be made to perform arbitrary computational actions:

We present fundamental progress on the computational universality of swarms of micro- or nano-scale robots in complex environments, controlled not by individual navigation, but by a uniform global, external force. More specifically, we consider a 2D grid world, in which all obstacles and robots are unit squares, and for each actuation, robots move maximally until they collide with an obstacle or another robot. The objective is to control robot motion within obstacles, design obstacles in order to achieve desired permutation of robots, and establish controlled interaction that is complex enough to allow arbitrary computations. In this video, we illustrate progress on all these challenges: we demonstrate NP-hardness of parallel navigation, we describe how to construct obstacles that allow arbitrary permutations, and we establish the necessary logic gates for performing arbitrary in-system computations.

[ SoCG 2015 ]

Ridgeback + UR5 = Omnidirectional Mobile Manipulation:

The most amazing part of that video was watching the Ridgeback and UR5 grab the tire without relying on any sensors whatsoever. Magic!

[ Clearpath Robotics ]

Thanks Ilia!

Either IHMC has kept themselves busy since the DRC Finals, or they’ve just gotten around to posting some old videos. But who cares, this is ATLAS doing cool stuff, like walking fast enough that it collapses from hydraulic fatigue (evidently that’s a thing).

[ IHMC ]

Building a prototype robot is a lot of work. Listen to the VP of engineering for Jibo explain to you just how much work it is:

[ Jibo ]

And now, an “Analysis of Deceptive Robot Motion,” featuring Herb Mugface:

Much robotics research explores how robots can clearly communicate true information. Here, we focus on the counterpart: communicating false information, or hiding information altogether – in one word, deception. Robot deception is useful in conveying intentionality, and in making games against the robot more engaging. We study robot deception in goal-directed motion, in which the robot is concealing its actual goal. We present an analysis of deceptive motion, starting with how humans would deceive, moving to a mathematical model that enables the robot to autonomously generate deceptive motion, and ending with a study on the implications of deceptive motion for human robot interactions.

Read the RSS 2014 paper, by Anca Dragan, Rachel Holladay, and Siddhartha Srinivasa from the Robotics Institute at Carnegie Mellon University, here.

I hadn’t heard of this “Bad Land: Road to Fury” movie before, but it’s got a robot in it that’s based on Big Dog:

According to Windmill’s visual effects supervisor, Ditch Doy, the director’s original intent was to use a fully-functioning robot. “ [Writer/Director Jake Paltrow] went to Boston Dynamics and tried for a couple of weeks to get a real robot for the film,” Doy revealed. “That just wasn’t realistic, but if Jake had had his way it would have been a completely robotic creation.”

Yeah, get in line, buddy.

[ Cinefex ]

How long can your MAV hover for? Not this long:

This is the flight test video which demonstrates 31 minutes of sustained controlled hover [from single battery charge] of the 45 gram high-efficiency quad-rotor helicopter developed at University of Maryland [UMD]. The video is sped-up 4 times, but we have included a timer which shows the real time. As far as we know this is the highest hover endurance by an electric powered hover-capable MAV in this weight category. The vehicle was designed and built over the last 3 years [2012 – 2015] by Justin Winslow [UMD], Prof. Moble Benedict [currently at Texas A&M], Dr. Vikram Hrishikeshavan [UMD] and Prof. Inderjit Chopra [UMD]. The test pilot was Dr. Derrick Yeo [UMD].

If you thought that Robot Town was somewhere near Boston, Pittsburgh, or in Silicon Valley, you’d be wrong.

The town Peccioli in Tuscany is also called the "Robo Town". Here one of ECHORD++'s three Robotics Innovation Facilities (RIFs) is located. The RIFs provide robotic software and hardware platforms, as well as profound knowledge for system integration. They offer easy and non-bureaucratic access to robotic equipment, as well as to experts to support bootstrapping. And what's probably the best part of it: everything is free of charge.

[ Peccioli RIF ]

Techi is an autonomous navigation service robot for hotels and hospitals consisting of an Adept Lynx mobile base with a customized upper bit from Techmetics:

“A first of its kind?” I dunno about that...

[ Techi ]

Mira is a desk companion that makes your life better one smile at a time. This project explores human robot interactivity and emotional intelligence. Currently Mira uses face tracking to interact with the users and loves playing the game “peek-a-boo”. As her understanding of the world and people's emotions get richer so will her ability to interact with people in a more meaningful way.

Mira is being developed by Alonso Martinez, a character technical director at Pixar Animation Studios.

Are we done yet? Almost! A little BiT more from DARPA BiT:

Dr. Phillip Alvelda, Program Manager in DARPA’s Biological Technologies Office (BTO), discusses the potential of next-generation neural interfaces to improve quality of life for people and revolutionize how we engage with machines. The talk was part of a two-day event held by BTO to bring together leading-edge technologists, start-ups, industry, and academic researchers to look at how advances in engineering and information sciences can be used to drive biology for technological advantage.

[ DARPA BiT ]

In this presentation from June 21, 2015, Dr. Paul Mahaffy, planetary scientist at the NASA Goddard Space Flight Center and instrument lead for the MAVEN NGIMS and the Mars Science Laboratory SAM instruments, focuses on current and past measurements of isotopes and volatile gases in the Martian surface and atmosphere during the final day of the 2015 MAVEN New Media Professional Development Workshop.

During the presentation and discussion, Dr. Mahaffy demonstrates the compelling case that NASA's Mars missions are making for long-lasting water on the surface of the Red Planet and what that means for potential past and present habitability. 

[ MAVEN ]

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