GigaOM » robots

How robots can do more in data centers and lower the costs of operating the cloud

Jordan Novet — Wed, 22 May 2013 16:58:26 +0000

Machines are giving us better and better suggestions for things to read, restaurants to eat at and people to date. Behind the curtains, some of the ways these services are being delivered are also being automated.

An article out Wednesday from Data Center Knowledge envisions the next few steps for automating operations inside the data centers. Robots can move literally higher up the stack than humans and still be safe, which means data center builders can build vertically instead of horizontally. That could bring better use of data center floor space.

If robots do all the work on the floor, lights might become unnecessary, and poof: just like that, a line item can be nixed from the budget. Deploying robots could also lead to less downtime, as they could act with more certainty than people when it comes to replacing a server or another hardware component.

Using robots to grab equipment is “becoming quite feasible,” and Google does it to get backup storage tapes, according to the article. Most gear isn’t really made for machines to handle, though, so this area might be in need of tinkering before it can get widely adopted.

The article also makes mention of unmanned data centers, including one operated by AOL. Apple revealed plans last year to build one of these facilities in Prineville, Ore., before saying it would expand the site to add data centers where some people would work. As more companies move in that direction, prices will drop, leading to further market penetration.

Despite this, the article suggests that data centers will still need administrators, so not everyone working inside data centers will lose their jobs as this wave of automation carries through — for now.

Meanwhile, data center admins can also optimize their facilities by changing out hardware and software to match use cases. Pat Gelsinger, CEO of VMware, will talk about his vision for the software-defined data center, and Andrew Feldman, general manager and corporate vice president of AMD, will talk about how companies can do these things at GigaOM’s Structure Conference in San Francisco on June 19.

With these sorts of upgrades, while the initial capital expenditures might be high, they could bring operating expenses down for public, private and hybrid cloud providers, resulting in price drops for customers in time.

Related research and analysis from GigaOM Pro:
Subscriber content. Sign up for a free trial.

How to make a less creepy robot? Simple, just add data

Stacey Higginbotham — Mon, 20 May 2013 21:22:00 +0000

Disney’s research arm has solved a problem that you probably didn’t even know robots have — their inability to accept objects from people in a natural way. The Disney Research team, working with funding from the International Center for Advanced Communication Technologies (interACT) at Carnegie Mellon and the University and Karlsruhe Institute of Technology (KIT), believe that robots who can’t naturally accept “handoffs” of objects from people are creepy. In a paper presented this month, Disney and its partners detailed how they used several motion-sensitive cameras, a database of gestures and some fancy algorithms to solve this handoff problem.

From the press release announcing the findings:

“If a robot just sticks out its hand blindly, or uses motions that look more robotic than human, a person might feel uneasy working with that robot or might question whether it is up to the task,” Katsu Yamane, Disney Research, Pittsburgh senior research scientist explained. “We assume human-like motions are more user-friendly because they are familiar.”

Despite the robot pictured on the Disney page touting this research looking like the mechanical, blue-haired skeleton that haunted my childhood nightmares, its attempts to grab the purse from the person do seem reactive to the human’s gestures, as opposed to the robot just sticking his arm out there and the person having to accommodate it. And that sort of naturalism will be important as we bring more robots into our homes and workplaces.

For example, an MIT group used a dancer’s motions to build a robotic bartender in a quest for naturalism — even though that robot doesn’t interact with people.

Today, designers try to endear robots to us with quirky noises (like R2D2) and maybe light displays or LED faces — anything to help anthropomorphize them. But as robots become more human-looking they can also become more sinister — achieving that same uncanny valley that Disney and other content companies have struggled with in animation. Remember the dead-eyed stars of the Polar Express that you probably couldn’t empathize with? The jerky movements of a home health robot might engender similar feelings — or worse — they may scare people.

Building the natural gestures of the Disney robot took the creation of a hierarchical gesture database that the robot can access as it detects the person passing something to it. In the Disney paper research, the robot is not only able to reach for the handbag, but when the human attempts a fake pass to the robot, the ~~blue-haired monstrosity~~ robot is able to adapt. From the release:

To enable a robot to access a library of human-to-human passing motions with the speed necessary for robot-human interaction, the researchers developed a hierarchical data structure. Using principal component analysis, the researchers first developed a rough estimate of the distribution of various motion samples. They then grouped samples of similar poses and organized them into a binary tree structure. With a series of “either/or” decisions, the robot can rapidly search this database, so it can recognize when the person initiates a handing motion and then refine its response as the person follows through.

Even if you don’t have an opinion on how naturally robots should move, this research brings home the awesome amount of work it takes to build computers and robots that mimic the capabilities of a person. Much like computer visualization, the science of robotic interaction takes a problem the size of a mountain and has to chip it down into grains of sand using a toothpick to find solutions. It’s a testament to human curiosity that people are willing to try.

Also, I expect Disney might be lured by the idea of natural-looking robots roaming its theme parks. My only question is would they be dressed up as characters or working the cash register at the gift stores.

Related research and analysis from GigaOM Pro:
Subscriber content. Sign up for a free trial.

Just add robots: MIT and Coke show off a smartphone-controlled bartender

Stacey Higginbotham — Mon, 13 May 2013 14:48:27 +0000

We could be saving the world, y’all. But instead researchers at MIT’s Senseable City Lab in Cambridge have worked with the The Coca-Cola Co. and Bacardi companies to deliver a smartphone-controlled mixologist that can make an infinite number of drinks (both alcoholic and non-alcoholic) with the touch of a few buttons. Which admittedly, is cool.

A device like this is both a mechanical poster-child showing what’s possible for the future of automation as well as an attempt to work out kinks on our way to personal robots that will help reduce drudgery and improve our lives. And yet, despite my love of cocktails and robots, this particular partnership rubs me the wrong way.

It’s called the Makr Shakr and the robot and smartphone app will be unveiled Wednesday at Google’s I/O conference. The apparatus has three robot arms, whose movements were modeled after those of an Italian ballet dancer Roberto Bolle, and access up to 100 different types of flavors (alcohols, fruit juice, syrups, etc.) You download the app and then use the app to create the mix of your choice.

MIT is making a big deal of being able to see what recipes people make and how the robot allows others to crowdsource winning combinations. For example, the release includes the following quote, which is presented without irony:

“The number of drink combinations is almost limitless,” says Yaniv Turgeman, project leader from MIT Senseable City Lab. “The magic moment will be watching the formation of a bottom-up bar culture, as we close the loop between ‘co-curating’ and ‘co-producing,’ in real time.”

To which I say, Bah humbug. Yes, a robot-dispensing bartender whose “arms” are modeled after a ballet dancer and are assigned movements by a professional choreographer (it’s Italian director and choreographer Marco Pelle if you must know) is clearly theater. But this is art used to package commerce in a form designed to appeal to the frenetic, internet-hardened consumers of today. While plenty of hackers and hipsters might feel immune to a TV spot or a billboard, they will flock to tell their friends about this robot, post videos of it online and in general look without skepticism at what the creation of this robot means.

And of course, people like me, will write articles about it. Because it is cool, and these sorts of endeavors are the future, of both the service industry and commerce. And while MIT is careful to note that this won’t replace your neighborhood bartender, that’s just idiotic.

In Austin, Briggo Coffee, is a robotic coffee-dispensing robot (it’s shaped like a giant cube) that allows you to order your cup of Joe on your smartphone on the way into work and pick it up form the cube at a set time. Let’s also not forget that in the heart of SoMa at Lemnos Labs, Momentum Machines is building a robot that makes hamburgers. And in Amazon’s warehouses robot pickers abound.

The Briggo coffee-making robot lives inside that cube.

Robots are cheaper at scale, and more efficient than people. For example, MIT boasts that Makr Shakr can not only mix drinks, but can also monitor alcohol consumption and blood alcohol levels. So while it may not wipe down glasses while listening to your troubles, the Makr Shakr has a lot going for it should anyone decide to turn it into a production robot.

So while this is whiz-bang cool it’s worth asking the tough questions such a partnership and device bring up. Is this really what we want our MIT graduates working on? And if it is, then what kinds of adjustments do we need to be making in society to prepare for the infusions of robots into the service sector. Perhaps folks can start discussing those things after they get their Makr Shakr-made drinks.

Related research and analysis from GigaOM Pro:
Subscriber content. Sign up for a free trial.

First in flight: Maryland professor’s robot bird good enough to fool the real thing

Amanda Alvarez — Wed, 01 May 2013 23:07:52 +0000

Satyandra Gupta apparently loves birds so much he decided to build one. His skills as a professor of mechanical engineering at the University of Maryland probably didn’t hurt in his quest, and this week he announced the Robo Raven is now a reality. The robotic avian can dive and roll and looks so realistic that other birds have attacked it in flight.

Developing the robot bird was a decidedly start-and-stop affair. Over the course of eight years, design flaws caused incapacitating crashes in each iteration of the robot. 2007 saw the first successful flight by a prototype with simultaneously flapping wings. By 2012, Gupta and colleagues had succeeded in developing a model that could flap its wings independently. For the robot, at least, simultaneous wing-flapping was a drawback. Engineering independent wing-flapping behavior was time-consuming, and also made the robot heavier.

The Robo Raven has two motors that are coupled to coordinate the movements between the two wings. It can be programmed with arbitrary flight patterns, as can be seen in the video below. To compensate for the additional weight of a bigger onboard battery and microcontroller, the robotics team used lightweight 3D printed parts for the body. Aerodynamic optimization allowed the Robo Raven to reproduce observed flight behavior of real birds.

Like quadcopter drones, the future of flapping wing micro air vehicles may lie in surveillance, or in just looking really really cool.

Related research and analysis from GigaOM Pro:
Subscriber content. Sign up for a free trial.

Baxter the Robot — now available for research

Barb Darrow — Thu, 25 Apr 2013 19:23:14 +0000

Remember Baxter the Robot? The $22,000 machine initially built for manufacturing applications can now be repurposed for research tasks, according to Rethink Robotics.

Baxter’s advantages lie in its ability to “learn” tasks from a person who walks it through a series of motions. The trainer need not be a computer programmer. And, because Baxter is sedentary — it doesn’t roll or walk around — there is no need for protective cages to separate it from human co-workers. For tasks that can be performed at a countertop or an assembly line, Baxter could be an ideal worker.

Rethink has said all along that Baxter will move on to other types of jobs, including, potentially, home healthcare. Last fall, Rethink CEO Rodney Brooks promised a Software Development Kit (SDK) that would enable Baxter to be refashioned for new jobs over time. “Our story is manufacturing, but there will be new software every two to three months with new capabilities,” Brooks, pictured above, told EMtech 2012 attendees in October. “Researchers will find places to use it that we wouldn’t have guessed.”

Here’s the scoop from Rethink’s web site:

“Baxter Research Robot is a $22,000 humanoid robot platform with two 7-axis arms, integrated cameras, sonar, torque sensors, and direct programming access via a standard ROS interface. It is entirely safe to operate around humans without safety cages, making it the perfect companion for late nights in the lab… with no extra pizza required.”

Related research and analysis from GigaOM Pro:
Subscriber content. Sign up for a free trial.

Researchers create cloud-based brain for robots

Derrick Harris — Mon, 11 Mar 2013 17:16:05 +0000

A group of European researchers has released the first version of a cloud computing platform for robots that will help them take advantage of powerful virtual resources. Essentially, they’re treating robots like any other device — desktop, tablet or mobile phone — running web applications, only robots can learn from each other and can do a lot more than just update screen displays.

The project, carried out by a team at ETH Zurich, is called RoboEarth and its linchpin is a cloud software platform called Rapyuta. The way it works is pretty simple at a high level: robots communicate with a cloud-based application platform that carries out computation tasks and connects to a cloud database full of information such as maps, images, language, as well as to other web services. The robots themselves are pretty much hardware terminals equipped with sensors and moving parts but limited on-board processing power or data storage.

This, of course, is an idea that has been with us since the mainframe computer and continues today via cloud computing and web and mobile applications. Why limit a device to its own physical capabilities when there’s an infinite (although, in the case of mainframes, not so much) expanse of computing power, memory, storage and data available in the ether? As long as the device has a strong internet connection, it doesn’t need a massive hard drive or the latest, greatest processor.

One thing RoboEarth does a little differently, though, is allow for databases that robots can update as they go about their business in different situations in different parts of the world. It’s machine learning, only in a much more literal sense: robots are actually learn from the experiences of other robots. That learning, in turn, should make them more useful to humans who won’t have to program them as thoroughly and perhaps can use the robots to perform a wider — and ever-expanding — variety of tasks.

Presumably, though, some statistical machine learning on the backend could make the robots even smarter as they generate more and more data and patterns begin taking shape. (We’ll be talking about unique ways to put machine learning to work at our Structure: Data conference next week in New York.)

The easy joke to make about this type of project is to say it’s the start of SkyNet and the rise of the machines, but that’s a bit of a stretch. After all, the machines themselves aren’t communicating with one another but, rather, with a centralized computing infrastructure operated by humans. It’s similar to IBM’s Watson system, which is really good at answering questions, but only as good as its information database and algorithms allow it to be.

If you’re curious to learn more about the promise and limitations of something like RoboEarth, Markus Waibel, one of the project’s researchers, has a great blog post explaining his vision of the project and where it fits into the greater ecosystem of web-based robotics.

Feature image courtesy of Shutterstock user Bruce Rolff.

Related research and analysis from GigaOM Pro:
Subscriber content. Sign up for a free trial.

How to train robots to work with people? Let them train people, too

Barb Darrow — Mon, 11 Feb 2013 18:56:41 +0000

Robots need to be taught by people to perform specific tasks. The traditional thinking has been: Train em’ up and let ‘em work while their human co-workers go on to do other things. Now, new research out of MIT posits that robotic and human “co-workers” can work better together by cross-training each other.

In heavy industrial applications like automotive manufacturing, robots perform tasks that are too big or too dangerous for human workers. In such jobs they are isolated from people for safety reasons. But what if human and robotic workers need to work together efficiently and safely in close proximity? That’s a problem Julie Shah, head of the Interactive Robotics Group at MIT’s Computer Science and Artificial Intelligence Lab (CSAIL) is attacking with new research. According to a statement Shaw made to MIT News:

“People aren’t robots, they don’t do things the same way every single time … so there is a mismatch between the way we program robots to perform tasks in exactly the same way each time and what we need them to do if they are going to work in concert with people.”

Traditionally, human trainers reward robots when they do the task correctly and provide negative feedback when they fail to do so. But military research has shown that a similar approach with people alone is inefficient and does not encourage teamwork. That led Shaw to look into other ways to train people to work together well and to see if those methods could also apply to mixed teams of robots and people.

Cross-training — in which team members switch roles on different days — gives members a better idea of how their individual work affects their co-workers. To bring robots into the fold, Shaw and Ph.D. candidate Stefanos Nikolaidis built a algorithm to teach devices how to learn from their role-swapping experiences.

They ended up tweaking existing reinforcement-learning algorithms to allow the robots to take in not only positive and negative rewards, but also other information pertinent to the job at hand. Their findings? Mixed teams that cross-trained were much more efficient than mixed teams that used the older interactive reward method.

The researchers discovered that the amount of time that robots and people that cross-trained were able to work concurrently rose 71 percent while concurrent work time by the other teams fell 41 percent. This is an important measure: If human team members have to wait for a robot to complete a task (or vice versa) before resuming work, that’s a lot of down time. Cross-trained robotic/human co-workers can work in tandem and get more done much more of the time than teams trained the old way.

Shah and Nikolaidis will present their research at the International Conference on Human-Robot Interaction in Tokyo next month. Check out the videos on their work . If this research pans out, look for more mixed workgroups on the factory floor and some pretty impressive productivity gains.

Related research and analysis from GigaOM Pro:
Subscriber content. Sign up for a free trial.

A robot that can help your grandma (or you) avoid the nursing home

Barb Darrow — Sun, 18 Nov 2012 15:30:51 +0000

As baby boomers age and retire, they’ll be confronted with a really tough choice: Pony up for extremely expensive assisted-care or full-care facilities or stay at home and rely on family or outside help as needed. Ask any aging person their preference, and you’ll see that the “age in home” option is the winner both in economic and psychological terms. The AARP estimates that 84 percent of respondents want to remain in their homes.

Willow Garage’s Personal Robot 2 with two friends.

The good news is that, relatively soon, it looks like there will be tech options — including in-home robots — that can help them stay put.

The technology is getting more adaptable and more affordable. Rethink Robotics sells Baxter, an industrial robot, for $22,000 now and, with the advent of a new software development kit, Baxter could be adapted for new markets. Rethink’s CTO and founder Rodney Brooks hopes that geriatric care will be one of them. Speaking at the Techonomy Conference last week, Brooks said:

“I think absolutely elder care is going to be an incredible pull on automation technology, because people want to stay in their homes longer, and the demographics is much more older people. This, by the way, is true in China now too. You know, a young Chinese person now has two parents and four grandparents who don’t have anyone else but that person.”

So there’s … going to be a real pull for how technology lets people be independent longer. And I’m hoping that someone will come up with some provocative things with Baxter for that.”

Research: Older Americans are receptive to robotic help

Technology is one thing. The willingness to use it is another. And new research out of Georgia Tech shows that acceptance of in-home robot help among older people may be growing. Researcher Cory-Ann Smarr, a PhD candidate in Engineering Psychology at the school, said a survey of 21 people between the ages of 65 and 93, showed them open to using robots for some tasks — but also very picky about which tasks those would be.

“They were fine with a robot reminding them to take their medications but not so fine about a robot telling them what medications to take,” Smarr told me in a recent interview.

Most of the respondents — who were shown a video of the Willow Garage PR-2 robot performing jobs — seemed pretty tech savvy already. Most were proficient users of cell phones, recordable and programmable devices like thermostats and coffee makers. Most (71.4 percent) said they used a computer so these respondents were not tech neophytes and may not be typical of older seniors in the broader population. Recent Pew Research on smartphone ownership found that just 11 percent of adults over the age of 65 have smartphones, compared to 34 percent of those aged 50 to 64 year. Still, it’s interesting that they did not shoot down the notion of a robot helper.

While there’s been talk of robots in the home for decades — Roomba helped blaze the trail — there’s really nothing out there that looks like a robotic butler or maid. But given the economics of aging, the option of staying in a home that’s paid for versus selling that home to pay for facility living is undoubtedly attractive. Since assisted care facilities can cost anywhere from $1,500 to $5,000 per month, the notion of buying a $22,000 robot — assisted at times by human helpers — might not be beyond the pale.

And more people will face that choice than ever. The most recent U.S. census data shows that the percentage of Americans over the age of 62 grew 21.1 percent between 2000 and 2010 — the second fastest growing demographic after the 45 to 64 year age group which grew 31.5 percent in that period. And, the AARP estimates that the number of US adults over the age of 65 will nearly double in the next 20 years.

Companies that can build adaptable and afforable robots to fill this growing need could be big, big winners going forward.

Related research and analysis from GigaOM Pro:
Subscriber content. Sign up for a free trial.

How can we help humans be more than robotic babysitters?

Stacey Higginbotham — Wed, 14 Nov 2012 16:17:07 +0000

“It turns out [boredom is] a much bigger problem in any system where a human is effectively babysitting the automation.” — Mary “Missy” Cummings, associate professor of aeronautics and astronautics at MIT

Cummings and researchers in MIT’s Humans and Automation Lab are investigating how people interact with automated systems, and in a study to be published in Interacting with Computers, she outlines how distractions may be the key to keeping the pilots focused.

The study and release announcing the results are here, but the topic is one that goes beyond the military. Automation is creeping into our lives in a variety of ways, from robots to help care for the elderly (but overseen by doctors) to factory floors where people no longer have a place. It’s being built into our scaled out computing systems, our telephone networks and any other place we can make it work.

Related research and analysis from GigaOM Pro:
Subscriber content. Sign up for a free trial.

Mail.ru chief Dmitry Grishin buys into Y Combinator grad Double Robotics

David Meyer — Fri, 21 Sep 2012 11:30:34 +0000

Last time we caught up with Mail.ru CEO Dmitry Grishin, he was launching a $25m fund to stimulate the robotics startup industry. It looks like he’s making good on that promise: on Friday Grishin Robotics announced a $250k investment in Y Combinator grad Double Robotics.

Double Robotics is one of a new generation of robotics firms that’s trying to bring the technology to the masses at a reasonable price. Its product, the ‘Double’, is a telepresence robot that uses a couple of iPads (one for control, one for the telepresence imagery) and costs $2,499, although it has a lower pre-order price of $1,999.

“[The] Double Robotics investment perfectly aligns to our strategy,” Grishin said. “It is a very consumer-oriented product with potential to fit a very wide range of applications and has already proven a strong consumer demand.”

Robots for the masses

I’d hesitate to call it a consumer product myself, but I can certainly see its application in many organizations. Telepresence has never been a cheap technology – though the cost is falling – and the Double’s use of iPads is a clever way to keep prices down.

And it seems to be working. In its first post-launch month, Double Robotics apparently picked up 600 pre-orders, equating to $1.2m, from buyers in 44 countries. According to Grishin, whose fund comprises his own cash, the first customers include 24 universities and 17 Fortune 500 companies.

That interest has led to scaling problems, which is where Grishin’s cash – the first to be parcelled out since the fund launched – comes in. With the first run already sold out, the fresh injection should allow the second set to be delivered early next year. It also looks like there may be further play between Grishin and Double in the future.

“We are thrilled to have Grishin Robotics and Dmitry Grishin, in particular, as our largest investor to date,” Double Robotics co-founder David Cann said. “After our public launch in August, we met with Grishin Robotics and were immediately impressed with their mission and deep knowledge of the robotics industry’s past mistakes and potential future. We look forward to working with Grishin Robotics in the years to come as we build our business.”

Here’s a reminder of what the Double looks like:

Related research and analysis from GigaOM Pro:
Subscriber content. Sign up for a free trial.