You Know You Want One: Personal Robots Are Coming, But Not Ready For You Yet : All Tech Considered A Silicon Valley company hopes to make a Jetsons-style robot a reality in your home. But the personal robotics industry has a few hurdles to jump before you can have your own robotic maid.

You Know You Want One: Personal Robots Are Coming, But Not Ready For You Yet

  • Download
  • <iframe src="" width="100%" height="290" frameborder="0" scrolling="no" title="NPR embedded audio player">
  • Transcript


From NPR News, this is ALL THINGS CONSIDERED. I'm Robert Siegel in Washington.


And I'm Melissa Block, hosting this week from NPR West in California.

We're focusing this week on the theme of innovation on the West Coast, innovation in industry, in agriculture and, today, a story about innovation in robotics. Think first about the robots you've seen on the screen.


BLOCK: Now, there's Rosie, the Jetson's maid with the red eyes and little silly apron.


BLOCK: And there's Wall-E, that cute Pixar robot, a boxy thing rolling around on all-terrain treads.


BLOCK: Well, that's a fantasy. We went to Silicon Valley to meet the robot of the real world at a company called Willow Garage in Menlo Park.


BLOCK: That's where we found Jake, 500 pounds, 4 feet 4 inches tall with a spine that stretches another foot.


BLOCK: He's got a white urethane skin, a flat head sporting an array of camera lenses, a laser scanner in his throat, jointed arms with grippers for hands, and he moves around on caster wheels.


BLOCK: Jake is one of Willow Garage's PR2s, that stands for personal robot. In robotics' jargon, the PR2 is a robot platform for experimentation and innovation. In other words, it's a vessel for developing new applications with Open Source software, so people all over the world can use it. And the idea is that the robots developed here will be in your home in the future.

So far, the PR2 has been programmed to fold towels, shoot pool and fetch a beer. Today, Jake is hard at work trying to a refrigerator door and take out a bottle of juice. It, he, Jake has raised his arms, right? So they're poised.

LEILA TAKAYAMA: If you look at the weight set up right now, I'm not sure he's going to make it, because his arms are not long enough...


TAKAYAMA: actually reach the handle. So we'll see what he does.

BLOCK: This is Leila Takayama. She's 31, a research scientist at Willow Garage. The company was started six years ago and it's considered one of the most exciting, influential players in the world of personal robotics, and that's were Leila Takayama comes in. Her specialty is human-robot interaction. She's got a background in psychology, so she observes people, and then figures out how to make robots more personable.

TAKAYAMA: Can they be in an environment with people? And that's not just, you know, don't kill the people?

BLOCK: So, Leila spends a lot of time thinking about how these robots should look. Less intimidating, for one.

TAKAYAMA: You could pose it as this big, big robot that's very scary and very capable and smart. Or you could pose it as something that's sort of more humble and smaller in personality. And so, you sort of, you know, tuck the arms in like this, as opposed to like lifting them big up like this.


BLOCK: Yup. Yup, here goes the robot. The arm is. The arm is poised. So back to Jake. He still is at the refrigerator, the laser scanner in his throat moving up and down sizing up that door. Leila ponders the robot and wonders if it could be more human-friendly.

TAKAYAMA: At the moment, it looks like it's going to punch the refrigerator but that's not what its trying to do.


BLOCK: Because the elbow is cocked back like that.

TAKAYAMA: Yes, exactly.


BLOCK: Leila Takayama says, wouldn't it be helpful if this robot could show us that it's thinking.

TAKAYAMA: Sometimes it's hard to tell, like, is it trying to get through the door. Is it trying to go down the hallway? Is it waiting for something? It's like it looks the same when its thinking really hard, as when it's doing nothing at all. Can we do something about that?

BLOCK: So, Leila has been working with an animator at Pixar to come up with ways to make these robots more humanoid. Maybe it's adding a gesture, like having the robots scratch its head to show that it's concentrating or adding sounds. Along with the Pixar's sound designer, she's developed a sound library.


TAKAYAMA: That one is one of my favorites for getting people's attention.


TAKAYAMA: So this is, Hey.


TAKAYAMA: And that's sad, sad.



TAKAYAMA: Yeah, I feel kind of bad for him.


TAKAYAMA: Ran out of battery.


TAKAYAMA: Please help.

BLOCK: Takayama says she'd love it if the robot could lean in and out from hips, not just roll back and forth. That would show expression and engagement. And another robot movement should love to see...

TAKAYAMA: A pop. And that's when you sort of just sort of sit up really fast. Right, and that's like I'm very interested or I'm curious, or I want attention. And that's dangerous to do in a real robot design because you're launching several hundred pounds of weight up really quickly.


BLOCK: Are there times when your perspective, as somebody with a background in psychology, butts up against the engineers...

TAKAYAMA: Oh yeah, absolutely.

BLOCK: Like I don't care if it's (unintelligible), I want it to do this.

TAKAYAMA: Yeah. So when we're designing PR2, a lot of the perception research folks wanted as many cameras as we could put on the head, right? Because then you've got lots and lots of data. And so, the original design had this sort of pile of cameras stacked high on top of each other, on top of the robots head. And that, to me, read as tarantula that's 500 pounds.


TAKAYAMA: And if you have a person who's not a roboticist interacting with a robot that looks like a gigantic tarantula, that's a problem. They're never going to get anywhere near it

BLOCK: Meanwhile, back at the fridge, Jake - ever patient - has inched toward the door. You can see dents where the robot's gripper hands has slammed into it trying to master this task. And finally, a breakthrough.

Well, look at that. Jake has opened the door - a little bit. Now, he's got to aim for the juice bottle.

TAKAYAMA: Oh, and now the fridge is opened, then it's slowly rotating down.

BLOCK: Spend time with robots, Leila Takayama says, and you start to see the world differently.

TAKAYAMA: I like that. Like, it's really nice to be able to appreciate. You know, the human hand is an amazing thing, especially after you spend the whole day watching this gripper try to grab a bottle...


TAKAYAMA: ...for hours. It's hard. It's really hard. And it's amazing that we can do what we do as humans. That's really fun. It's also neat to see just how people react to these things. Right? I get used to being around roboticists who are sort of bored with robots, because we see them every day. But when I get to bring in people from outside of the building, that's really interesting to me, because these are new technologies and we've got old brains.

BLOCK: What's the big picture here, Leila, do you think 10 years from now, 15 years from now, how different is your expectation going to be of what a robot should be doing?

TAKAYAMA: I want robots to be unremarkable. I hope that in 15 years we'll be at a place where people don't say: Oh, look, it's a robot, and then go right over to it. You know, it should just - they just should be there. Hopefully, they're so useful and so just sort of faded into the background that we don't notice that they're there all the time.


BLOCK: Whoop.

TAKAYAMA: Woo, so the...


TAKAYAMA: (Unintelligible) made a dive for something.

BLOCK: But it missed. And 40 minutes into Jake's refrigerator venture, his human handler across the room calls it quits. Anybody waiting for that juice would surely have given up by now, and the food in the open fridge would be headed for spoilage.

TAKAYAMA: Jake has almost successfully close the refrigerator door.

BLOCK: But never did get his juice.



TAKAYAMA: He did it.


TAKAYAMA: When he just shut the refrigerator door, which becomes a very exciting event...


TAKAYAMA: ...when you're a robot.

BLOCK: That's Leila Takayama working on human-robot interaction at Willow Garage in Menlo Park.

Now, Willow Garage has given away 11 of these PR2 robots, each worth $400,000, to research institutions. They're being programmed to clear a table, put away groceries, empty a dishwasher, and help elderly and disabled people with home care.

Now, clearly the PR2s have a long way to go, but their operating system, the Open Source software, is already transforming robotics. It's free for anyone to use, improve and share. The idea being that big advances in robotics won't come if people are starting from square one every time. So, what does the future look like for personal robots?

Ryan Calo follows the industry at the Center for Internet and Society at Stanford Law School. Ryan Calo, welcome to the program.

RYAN CALO: Thanks for having me.

BLOCK: And what about that future? How close are we to having robots in the home?

CALO: Well, there's this great quote by the science fiction writer William Gibson who says that the future is already here, it's just not evenly distributed yet. And so, the sense in which there are many kinds of robots already in the home and increasingly in hospitals and even small businesses.

But, you know, we're a few years out. I mean, a couple of years ago, Bill Gates wrote an op-ed in Scientific American, called "A Robot in Every Home," where he said that we are in a stage today in personal robotics that we were with personal computers in the '70s.

And so, you know, if you think about that, the Apple II didn't come out for a few years. It was, you know, early '80s. I think we are about five years away from the cusp of a really serious personal robotics industry.

BLOCK: A robot in every home, you think?

CALO: Eventually, right? I mean, so I think what it really needs to do is pass the neighbor test. You buy this robot and it gets delivered in a big box and your neighbor comes over and goes: Why did you just spend all that money on a robot? You've got to be able to answer your neighbor and say, well, I need it to walk my dog or I need it to wash the windows.

But once you get past that initial hurdle, it's going to be like a mobile phone, where you just constantly come across new apps that are useful to you. And then everybody wants one of these things because of the potential that they have.

BLOCK: If I understand this right - how the PR2 works - basically, the operating system, which is given to anybody who wants it for free - they say: OK, this is how this robot works, this is how it moves, this is how it's manipulated. It's kind of the nerve center, the brain of the robot. And then anybody else can take that and program it to do all kinds of things. Is that the idea?

That's right. And what's great about having a common ROS or robot operating system is that when you write code for that system, it can be run on any machine that runs ROS. And so, it's a real boon to the entire robotics industry to have this common language.

Well, when you think of the places where robots are used all the time, say, in industry, on assembly lines - why does it seem that home use, personal robotics, is so far behind?

That's a great question. So, the history of robots has been that they have been domain-specific, meaning that they're operating within a particular environment that you can control the parameters, right? And they're extremely good at that. Robots are very, very good at doing the same thing over and over again very fast and very accurately.

The challenge of personal and service robots is that they're not domain-specific. They have to negotiate all kinds of different environments. I mean, think about how different your friends' houses are. And the second thing is, you know, the safety considerations, right? I mean, if you go onto the floor of a factory, these large robots have a zone around them that you're not allowed to pass within because it's dangerous. Sometimes, they're actually physically fenced off, whereas the notion of co-robotics is the idea that robots will be working alongside us or helping us in our homes. And that's a real challenge.

You know, this past year, the Obama Administration committed about $70 million to the study of co-robotics, again, because the challenges are so serious.

We heard Leila Takayama in our story say that she wants robots to be unremarkable, right? How far away are we from that, do you think? When will that point come?

I think we're far from that point. You know, that said, there are also plenty of places where robotics is absolutely mainstream. I mean, chances are that if you've purchased something from Amazon, it was a robot that went and retrieved that item from the warehouse. So, for certain people, surgeons, those who are operating eCommerce, they already are part of the background and they already are boring. Right? It's just that, for the rest of us who don't encounter them on a daily basis, they're still going to be exciting.

Well, Ryan Calo, it's great to talk to you. Thank you.

Thank you.

Ryan Calo follows robotics at the Center for Internet and Society at Standford. You can watch a PR2 robot designed by Willow Garage folding laundry. That video is at


BLOCK: You're listening to ALL THINGS CONSIDERED from NPR News.

Copyright © 2012 NPR. All rights reserved. Visit our website terms of use and permissions pages at for further information.

NPR transcripts are created on a rush deadline by Verb8tm, Inc., an NPR contractor, and produced using a proprietary transcription process developed with NPR. This text may not be in its final form and may be updated or revised in the future. Accuracy and availability may vary. The authoritative record of NPR’s programming is the audio record.