TR Editors' blog

Just What Soldiers Need: A Bigger Robotic Dog

A next-generation robotic dog will help soldiers.

Kristina Grifantini 09/30/2011

  • 6 Comments

After BigDog, LittleDog, and Petman, we now have a glimpse of the newest intimidating robot from Boston Dynamics: AlphaDog.

In the video below, you can see an AlphaDog prototype trot along like a mechanical horse. It smoothly steps over loose rocks and a pipe, rebounds after a two-person shove, and rights itself when it's on its side. The technology is an evolution of BigDog, which could catch itself when skidding across ice and carry 340 pounds.

This motorized mule (officially dubbed Legged Squad Support Systems, or LS3) will be able to carry 400 pounds nonstop for up to 24 hours (or 20 miles) over rugged terrain. This could be a great boon to combat soldiers, who have to lug over a hundred pounds of payload. And rather than need a human driver, AlphaDog will use computer vision to follow a squadron leader on its own, or GPS to go to a predetermined location, according to Boston Dynamics.

The first completed version, funded by DARPA and the US Marine Corps, is aimed for 2012.

A Robot With Your Face

A telepresence robot due on sale next year hopes to do a better job of being you than any previous bot.

Tom Simonite 06/23/2011

  • 2 Comments
With the 21st century well under way and no signs of teleportation becoming possible, telepresence robots are our best chance at instantly being somewhere else. The pitch goes that you can jump onto (into?) your desktop computer and instantly move around, see and speak in a distant location.

Two such robots are already on the market--from Anybots and VGo--and now a third is set to join them. A robot research lab Willow Garage has spun off an independent company, Suitable Technologies, to develop its prototype telepresence robot Texai into a product.

It's slated to go on sale next year and sets out to solve a major problem as seen with the two robots already on the market: while a person inhabiting an Anybot or VGo robot gets a good(ish) view of their prosthetic body's surroundings and the people around it, those people don't get a good view of the operator's face.

Anybots' robot displays only a still photo of the current user, while VGo's machines have a very small, low resolution screen about four feet off the ground. "Those are really spy bots," Steve Cousins told me when I visited Willow Garage yesterday, pointing out that the people you're interacting with can't see you very well. Texai's big selling point over the competition will be that a user's face is clearly visible to the people his robot-double interacts with, enabling true, two-way communication, said Cousins.

It certainly seems plausible that this would make interacting via a robot a smoother experience. In my experience using a VGo to work in Technology Review's Massachusetts HQ from California, these machines struggle to meet the high expectations placed on something trying to fill the role of a person, as I noted in this review:

"My robot body could do some of the basic things I would do in person: move around the office to talk and listen, see and be seen. But it couldn't do enough. In a group conversation, I would clumsily spin around attempting to take in the voices and body language outside my narrow range of vision. When I walked alongside people, I sometimes blundered into furniture, or neglected to turn when they did. Coworkers were tolerant at first, but they got frustrated with my mistakes."

Perhaps if my distant colleagues had been able to see my facial expressions clearly the experience would have been easier for all. (Read about my experience.) But filling a big screen requires a big picture, which means bigger bandwidth. Anybots' founder told me they decided not to add operator video to their robots because Internet connections just aren't reliable enough to flawlessly send high quality video in two directions as well as a robot's commands.

Willow Garage has tested Texai extensively, with one of their engineers commuting to their office using one for nearly a year now. But I'm guessing their broadband connection is higher quality than in most homes and businesses. As I found out, connection woes are much more painful when they afflict your (robot) body, not just your Skype call.

CES: Robot with an iPad for a Brain

Is there anything Apple's tablet can't do?

Kristina Grifantini 01/07/2011

  • 1 Comment

Yesterday, at the 2011 Consumer Electronics Show, iRobot presented its latest prototype—a mobile robot that uses a tablet computer for its brain.

There have been plenty of other telepresence robots, ranging from video screens on wheels to strange, blob-like prototypes. The new robot, called AVA (for avatar), is an interesting twist, combining telepresence with autonomous control. A user can either operate Ava remotely, for a video conference perhaps, or let the robot maneuver around on its own, carrying out tasks assigned to it through the tablet acting as its head.

AVA is equipped with microphones, speakers, and a host of sensors, including two PrimeSense motion sensors (the same ones used in the Xbox Kinect), as well as sonar and laser rangefinders for mapping. It builds a map of its surroundings and displays it on the iPad. A user can touch a spot on the map to tell the robot where to go.

AVA could open up robotic programming to a host of people, potentially leading to interesting new applications.

PC Mag's Lance Ulanoff writes:

Angle explained that the inclusion of these mobile devices means that anyone who can program an iPad or Android app can now program a robot. Some more obvious uses include meeting telepresence, home and office security, material handling and grabbing you a beer from your refrigerator and bringing it back to you.

To see the robot in action, check out PCMag's video interview below.

About

Insights, opinions, and our editors' analysis of the latest in emerging technologies.

Subscribe to the TR Editors' blog RSS Feed

Advertisement
Advertisement

Facebook

Advertisement