The BBC Blogs - Spaceman
« Previous | Main | Next »

One giant leap for 'tin mankind'

Jonathan Amos | 19:58 UK time, Friday, 5 November 2010

It's not just the Discovery shuttle's six human astronauts who face an extra three-week wait to get into orbit. Their mechanical passenger is clenching his fists in frustration, also


Nasa has big plans for R2 but first the humanoid robot has to "earn its stripes" 

Packed in a box in the back of the orbiter is R2, the first human-like robot to be sent into space.

The robonaut is the product of 15 years' research in Nasa and General Motors.

In his current guise, he is just a head, arms, and a torso mounted on a pedestal. But the plan eventually is to give R2 some legs to let him move around the station. And in a couple of years, he'll also get a body upgrade that should significantly advance his capabilities.

The expectation is that before the decade is out, this robot will be clambering about on the outside of the space station, assisting astronauts on a spacewalk.

How realistic the computer graphics of this prospect really are, I don't know. The video artist clearly was a fan of the Boba Fett bounty-hunter from Star Wars. The visors are uncannily similar.

But, anyway - first things first. In R2's initial year in orbit, he will be confined to the station's Destiny lab, working away at a taskboard placed in front of him.

He has to demonstrate the full range of his dextrous capabilities, picking up and moving items, throwing latches and switches, and handling soft materials like cloth that have traditionally been problematic for robotic systems.

R2 also has to prove he is safe to have around. Rob Ambrose is the acting chief of the automation, simulation and robotics division in the engineering department at Nasa's Johnson Space Center:

"Superficially R2 is soft, padded and safe. Mechanically it is soft as well with springs. But deeper than that, it has three levels of electrical sensors and software, which can in each of the three levels identify forces that are being felt by the robot and decide if those forces are safe or not and stop the robot, if necessary. Those three levels of safety were essential in convincing our payload safety review panel that this robot would be safe to work adjacent to astronauts inside the space station."
R2 description


You've probably seen pictures of astronauts in the weightless surroundings of the station picking up fridge-sized metal boxes and moving them around as if they're made of cardboard.

They're not, of course; and a 100kg of box can do a quite a bit of harm if it's not handled properly. R2's behaviour has to be predictable but sufficiently reactive that he doesn't persist with a task where to do so might result in the robot damaging equipment or, worse still, crushing an astronaut.

On the ground, you see this reactive behaviour when someone walks into the robot's path. If his arms collide with an unexpected object, they immediately pause - they don't keep pushing regardless of the obstruction. But that's on Earth; how will these types of systems really work in space where the gravity loads are so different?

R2 will help engineers understand this sort of thing.

Certainly, the future looks bright for human-like robots. They will not only work alongside astronauts as humans push out across the Solar System, it is highly likely they will lead the way.

Who'd like to bet the first thing with two legs to walk on Mars will be mechanical, not human?

And, surely, that's particularly true of the more hostile locations one would eventually want to visit, like those fascinating moons of Jupiter and Saturn.

We wouldn't risk sending a human down to the surface of Europa with its unbelievably harsh radiation environment; we'd send a robot with legs to climb over the ice boulders and to use its dextrous hands to pick objects of interest. Rob Ambrose again:

"There are really three phases of exploration. The first phase is where the robots are really out ahead of the humans. The second phase is when the humans are at the same site as the robots, working alongside them. Typically these two phases will be short - of the order of weeks or months. And then, in between human crews going back to that site or never going back to that site, the robots would be left behind as caretakers, to run longer-term experiments."

Feel free to recall you favourite sci-fi robots. I still have a major soft spot for Huey, Dewey and Louie from Silent Running. I think that's because I have dogs and those little guys were like Bruce Dern's pets. Humans project personalities on to animals and objects. That's what we do. It will be no different for robots in space.

"At this point we have no plans for R2 to come back to Earth. I can't say what the future will hold. It may ride the station into the Pacific, but we don't know where station might go someday. That chapter in robonaut's life has not been written yet. We might take R2 beyond low-Earth. It's got to earn its stripes first."

R2 is currently discovering the frustration all humans have got used to with shuttle launches - delays due to technical glitches. Clive Simpson, the editor of Spaceflight magazine, is one of many who've been waiting around at Kennedy this week for Discovery to get off the ground.

He found R2's "brother" trying to pass the time and sent me a picture.  


R2's brother with a little light reading as he waits for the launch



or register to comment.

  • 1. At 11:23pm on 05 Nov 2010, callisto wrote:

    And on the 8th Day?

    Complain about this comment

  • 2. At 00:45am on 06 Nov 2010, Ricardo Wensleydale wrote:

    "R2's behaviour has to be predictable but sufficiently reactive that he doesn't persist with a task where to do so might result in the robot damaging equipment or, worse still, crushing an astronaut."

    Perhaps some kind of hardwired rules or "laws" would be in order....and don't forget to include allowing harm through inaction. Azimov would be proud.

    Complain about this comment

  • 3. At 08:02am on 06 Nov 2010, MarrsAttax wrote:

    "Open the pod bay doors please R2"

    Complain about this comment

  • 4. At 08:41am on 06 Nov 2010, The Realist wrote:

    The arms are capable of lifing 10kg.

    That robot will not be crushing anyone in space, in fact... it would get hammered by human, in Zero G or not!

    It is best to keep it this way to be honest until we have significantly improved in robotics.

    Complain about this comment

  • 5. At 08:54am on 06 Nov 2010, chrisk50 wrote:

    In the words of David Bowie

    Homo sapiens have outgrown their use.

    Exciting times, now it looks possible to extend journeys outside our galaxy, with robots tending the needs of humans in suspended life pods. Maybe not in my lifetime, but once robots can move and think for themselves they can be reproduced, even manufactured by themselves.

    Back to my original statement - Is there "life on Mars"?

    Complain about this comment

  • 6. At 08:57am on 06 Nov 2010, sensiblegrannie wrote:

    Can a robot suffer from stress from a lack of meaningful interaction? It will be interesting to see if the robot is as 'machine-like' as we would comfortably like to believe. Robot rights.

    Complain about this comment

  • 7. At 09:30am on 06 Nov 2010, Robonaut2 wrote:

    I don't wanna go! They didn't tell me that I'm gonna burn up when the space station comes down, see me on Facebook - [Personal details removed by Moderator]

    Complain about this comment

  • 8. At 10:42am on 06 Nov 2010, godofwine wrote:

    Looks more like a C3 to me, than an R2.

    Complain about this comment

  • 9. At 10:53am on 06 Nov 2010, Abs wrote:

    I find it interesting that the robot is given a gender and referred to as 'he'. Is this the correspondent's addition, or has it come from NASA? I wonder whether he's a white or a black male...

    Complain about this comment

  • 10. At 1:04pm on 06 Nov 2010, BluesBerry wrote:

    One giant leap for 'tin mankind', and I don't like it.
    Anyone who is totally comfortable with this innovation could not have seen the movie "2001 A Space Odyssey".
    Is it nicknamed, little "Hal".
    My expectation is that before the decade is out, this robot will be clambering about on the outside of the space station, creating havoc, and while the sapce shipo is undergoing destruction, endangering, if not killing, human beings.
    But, anyway - first things first.
    How much did R2 cost?
    Has R2 been earth-tested and proven safe?
    On the ground apparently, you see this "pause" behaviour when someone walks into the robot's path. If his arms collide with an unexpected object, they immediately pause - they don't keep pushing regardless of the obstruction. But that's on Earth; how will these types of systems really work in space where the gravity loads are so different?
    It disturbs me greatly that some men (really overgrown children paid large governmental salaries) get play with dangerous toys while the rest of us try to make a living. What next, the robot warrior - undefeatable, able to move a fridge like it's cardboard, able to leap tall building, faster than a speeding bullet...Able to combact the enemy with total and annihilating force?
    I don't care whether R2 is "soft, padded and safe". I care if R2 is "soft, padded and safe to be around". Would you want to carry out your astronaut duties with this robot adjacent to you, inside the space station?
    There is not a robot yet developed that makes our future look bright because in order to make out future bright they would have to have intellect and compassion. What good is a robot to todays' military people that has intellect and compassion?
    Who'd like to bet the first thing with two legs to walk on Mars will be mechanical, not human? I'll take that bet - against.
    Nice picture R2 "reading". Let's give it a comprehension test so we know that it can read a complex manual and follow instructions in space.

    Complain about this comment

  • 11. At 3:47pm on 06 Nov 2010, smallvizier wrote:

    @Abs, he's clearly white - just look at the picture! But I know what you mean, we're going to have to figure out how to talk about (and to!) human-like robots over the next few decades - and it will become increasingly important.

    I find it hard to believe R2 might eventually burn up with the space station. Most likely by the time the station is decommisioned, R2 will be a minor celebrity, or at least a heritage item. He'll end his days in a museum or maybe even giving talks for NASA...

    Complain about this comment

  • 12. At 5:05pm on 06 Nov 2010, No Worries wrote:

    10 BluesBerry...I think you're right. NASA have obviously misunderstood the technology or the ability of the robot to develop its own personality. Without the proper safety protocols it will probably kill all the astronauts on the space station. After it comes crashing back to earth R2 will probably be the only surviving piece left from the charred remains. Mad at our attempts to kill him he will start destroying all of mankind's achievements. Having also (through some kind of technology swap with other lifeforms) developed a force field that our weaponry cannot penetrate, he will become indestructible. Yes, we've certainly sealed our doom here.

    Alternatively, well done NASA and good luck!

    Complain about this comment

  • 13. At 8:24pm on 06 Nov 2010, friedemann wrote:

    Interesting bit of technology. But me thinks too many people have been reading doomsday books. "Attack of the Robots" It is a fairly big jump to give these toys a "personality" or "have the ability to have a desire". In order to have a killing insticnt, you need to "desire or want". So far all they are are computers with arms that follow the orders of the man-bot.
    We send robots to the bottom of the sea to close a oil leak (because the pressures are too great for a diver, and he would be crushed in mere seconds. Space is just as forboding. With the lack of air, how long does a man last? Yeah, just as long.
    What will get to Mars first? "Probes have been there". Already they have told us that there is no air worth talking about, water is not apparent and it is freaking cold.
    So after having this look over of the barren cheeseless real-estate, why are we interested in going there? To learn something we already knew?
    The "Garden of Eden is here" and we have to clue into that fact that we are messing it up. We ain't gonna get a second chance.

    Complain about this comment

  • 14. At 9:23pm on 06 Nov 2010, Robert Lucien wrote:

    R2's certainly a very impressive piece of tech, but I must say some of the responses here are rather humourous. Firstly this is NASA one of the worlds most paranoid organizations when it comes to health and safety, R2 is part of a very slow careful and meticulous development that has been running since 2002 or the mid 90's.
    Whats more hes not actually an independent robot, he's mainly controlled using 'telepresence'. As far as I know he's really controlled by a man in a VR suit with a headset - like the guys in Lawnmower man, or by the more clumsy laptop type interfaces. I think he/it has autonomy but only for very simple repetitive tasks and it certainly doesn't have AI or autonomy.

    R2's real purpose is to allow astronauts to work outside without going outside. This should allow them to do far longer and more complex tasks, with higher dexterity and precision. Spacesuits are bulky, heavy and clumsy, and above all slow (inc hours of prep before and after use) - and of course going outside is inherently dangerous. What R2 really does is simulate the limbs and joints and movements of a human very precisely, something thats actually very complex and difficult to do.

    Complain about this comment

  • 15. At 9:33pm on 06 Nov 2010, Bob Shaw wrote:

    Let me be the first to welcome our new overlords!

    Complain about this comment

  • 16. At 01:31am on 07 Nov 2010, Alex__T wrote:

    @Abs: "I find it interesting that the robot is given a gender and referred to as 'he'. Is this the correspondent's addition, or has it come from NASA? I wonder whether he's a white or a black male..."

    It's clearly a light hearted reference to the character R2-D2 from Star Wars, often referred to in the film simply as R2, who is given a male persona as far as is possible with a fictional robot.

    If you have a particular issue with this I suggest you take it up with George Lucas rather than the reporters at the BBC or the engineers which came up with this rather impressive piece of engineering.

    Alternatively you could just get over yourself and write something constructive for us to read...

    Complain about this comment

  • 17. At 12:46pm on 07 Nov 2010, Locke wrote:

    It could possibly be a danger to astronauts, but does it really make their job that much more dangerous? Also, their not being sexist when they call it he. If you see it walking down the street and can't call it it, what would you most likely call it? Everybody who thinks that this is the first chapter of Robot Apocalypse needs to stop watching movies and get educated.

    Complain about this comment

  • 18. At 2:44pm on 07 Nov 2010, Papa Ray wrote:

    Ah yes...a new thing, a new innovation, idea, invention.

    Actually not so new, we have been living and working with robots of many different kinds and functions for years. Have they killed any humans, injured them, caused them grief? Well yes, I'm sure that they have. Just like previous innovations, ideas and inventions have and will continue to do.

    Like the flint rock. It sure caused a lot of fires and I'm sure that the early humans were injured or even killed by this innovation.

    Like the first man made structures that were made of stone. I'm sure that many collapsed and crushed humans causing strife and fear.

    Like the first wheels. I'm sure that many were killed or injured by run away contrivances that had the dangerous wheel invention.

    Like the first engines, water powered, wind powered or even human powered. Even before the engines powered by hydrocarbons. I'm very sure that many thousands were injured or killed by these new fangled inventions.

    Then comes the big list of new innovations and inventions by mankind. Too long to list here, but we all (or at least the educated, informed of us) know that those inventions and innovations have killed millions upon millions and injured many more than that, over the course of human history.

    Do you see my point by now? If not you need to be reading nursery rhymes and doing watercolors and staying home in your little bubble.

    Then (and there) you can safely continue your life of contentment and never fear...

    Until of course the ROBOTS ATTACK !!!!

    Papa Ray

    Complain about this comment

  • 19. At 4:20pm on 07 Nov 2010, inconyito wrote:

    Oh, but Silent Running was SO BAD. And that's coming with someone who basically sympathized with it.

    I'd be most interested in the real current level of R2's autonomy, since a previous poster called it into question.

    Complain about this comment

  • 20. At 8:53pm on 07 Nov 2010, Jonathan Amos wrote:

    @Robert, @incoyyito. The robonaut development started out with the teleoperation concept in mind, but R2's sophisticated programs are actually run from a laptop, either on the station or from the ground.
    @Abs, @Smallvizier. He/She can be anything you want. Like I say, we humans project personalities on to objects. Pick your flavour .

    Complain about this comment

  • 21. At 01:26am on 08 Nov 2010, NL101 wrote:

    BluesBerry wrote:
    "My expectation is that before the decade is out, this robot will be clambering about on the outside of the space station, creating havoc, and while the sapce shipo is undergoing destruction, endangering, if not killing, human beings."

    We definitively don't need robots for all that, BluesBerry. We are already doing a bang-up job as it is unless you don't keep up with the horrific news from around the world. Perhaps these wonderful machines will teach us our own almost lost and almost gone human values for us to re-learn to take care of each other.

    Complain about this comment

  • 22. At 04:03am on 08 Nov 2010, John Hunt wrote:

    Have the R2 inside for a year flipping switched before it can be trusted to be outside? I don't think that 10 lbs of pressure is going to damage anything.

    NASA is too risk adverse. This is a big reason why there are so many delays and cost overruns.

    Project M wants to send R2 to the Moon. That would be great. But instead of it picking up rocks and making profound gestures, it ought to demonstrate that a teleoperated robonaut can make detailed repairs (on itself for example), set up solar panels and other simulated equipment, and inflate a greenhouse. By doing so, it could show that a low-cost, no-human phase could set up a Lunar Ice To LEO (LITL) mining system which would completely open up the solar system to humanity.

    Complain about this comment

  • 23. At 07:19am on 08 Nov 2010, ELA wrote:

    Why don't they just admit that they never went to the Moon? You know, this robot-tech-thing happening right now is in direct response to the rapid development of space-tech in China, who are "threatening" to go to the Moon with humans relatively soon in China's hopes and estimation, and NASA and America are SHIT-scared that the world is going to find out that there's nothing on the Moon that NASA left there supposedly, because THEY NEVER WENT. So they're going to fly this robot up there real quick with a small shuttle and drop it on the Moon and leave a flag there and some other remnants of the "era" to say that we went. Of course they'll just jettison the robot and its attachments and rocket parts out into space as soon as the mission is complete, and then show us ANOTHER COPY R2 at the Shuttle working, to make us think that we didn't lose one out to space.
    And then one day soon some Alien ship is going to catch that robot and fling it back to earth and say "did you lose something?"

    Complain about this comment

  • 24. At 11:41am on 08 Nov 2010, tmsanders wrote:

    In light of the advances in technology represented by the development of R2, it is really a shame to read posts that bring up and dwell on such irrelevant matters as the gender or race of a machine. The robot is referred to as "he" in the tradition of any gender-neutral object defaulting to male. The reference "it" is much too cold a label for such humanoid beings. As for race, anyone who brings this up in the context of advanced robotic technology has a serious problem and should seek professional counseling.

    Complain about this comment

  • 25. At 10:56pm on 08 Nov 2010, Treefrog wrote:

    A pity they couldn't find some robot p0rn to pose him with ;-)

    Complain about this comment

  • 26. At 01:28am on 09 Nov 2010, Granten wrote:

    In re. to RicardoWensleydale: Unfortunately that's actually not possible. We can't really program a robot to never harm a human because first we need to make sure it can definitely recognize a 'human' when one appears. Then we need to make sure that the robot can understand what 'harming' a human would be defined as. Both of those things seem obvious to humans but that's only because our brains have an incredible amount of data and the ability to process it well enough that we don't have to consciously ask ourselves "what would harming a human be like?"
    Also the Three Laws aren't remotely extensive enough to handle the various hypothetical situations that would result.

    On the robot itself, frankly I am not impressed. Why would they do something so foolish as to give it a humanoid design? There's a reason that the robots on Mars have wheels and a design that keeps them closer to the ground. Next NASA will want to give it eyes that can only see the same range as a human.

    Complain about this comment

  • 27. At 1:46pm on 09 Nov 2010, matt prince wrote:

    I wonder when they will buld C3PO

    Complain about this comment

  • 28. At 1:47pm on 11 Nov 2010, Fedya wrote:

    Hate to kill the enthusiasm, but it is a serious waste of money! These guys at NASA are playing around with remote control toys. Most important part of a Robot is the brain which unfortunately won't be around for another 20 years or so. Intel predicts that it will be able to keep-up with the Moor's law (doubling of transistors every two years) for another 15-20 years which should bring technology to a point of possibly emulating human brain. Terabytes of RAM, thousands of parallel cores may mimic 100 billion neurons which the human brain is assumed to have. I wish there would be advisers that could tell a hoax project from a real useful thing like more money spent on research of neural networks.

    Complain about this comment

  • 29. At 2:03pm on 12 Nov 2010, AlexTunbridge wrote:

    People are overestimating the intelligence of this robot. He's as likely to go AWOL as my PC (what's that flashing red light ...?)

    What is amazing is the dexterity. Hopefully, Robonaut has the capability for "remote control" (eventually from a call centre in Mumbai :) )as well as autonomous control. As long as Robonaut is less than a second away, there is very little need for autonomous control - he can be fully controlled from a team on the ground.

    This could come in handy for tasks outside. In a vacuum, the space suit has pressurised air and tends to balloon out. This means it's an effort to clench a fist, and difficult even to hold a spanner. Robonaut should suffer none of these problems.

    To make gloves easier to use, the space suits can operate at very low pressure. Then however the astronauts need to decompress before hand. This makes the whole process of space walking time consuming and awkward. Eventually, Robonaut should live in a box, outside the pressurised modules.

    Complain about this comment

  • 30. At 9:57pm on 13 Nov 2010, Antryg wrote:

    I for one would like to take this opportunity to welcome the arrival of our robot overlords...

    Complain about this comment

  • 31. At 00:22am on 14 Nov 2010, Ian Mc wrote:

    Two points; one, I don't think this is a tool that you are going to leave out in the shed as suggested above (#29). Space and humans don't mix well, as noted out by freiedman (#13), it's like trying to house a jellyfish in a frying pan. To do it requires a lot of environmental control on top of which lots of bulky food and water are required then the thing will insist on sleeping. R2 has none of these drawbacks, it can be operated/work round the clock by teams of drivers on the ground and will keep going as long as the solar panels can supply it with power which is generated on site and doesn't need shipping up at $10,000/kg. I work with robotic systems and for repetitive tasks they outperform humans, more reproducible, more accurate, don't get bored/make mistakes, etc. There will always be a need for a human presence on the Space Station but if this system proves itself humans may not make up the majority of the pairs of hands operating on the system.

    Two, the robot looks humanoid because it is quite difficult for people to drive something through a telepresence rig which has eight arms with six joints (OK if you are spider though, but the spiders at NASA seem to be, very literally, spaced out).

    Complain about this comment

  • 32. At 11:43pm on 18 Nov 2010, Simonm wrote:

    At 08:02am on 06 Nov 2010, MarrsAttax wrote:
    "Open the pod bay doors please R2"

    Careful with this one guys, Apple claims to have copyrighted the term 'Pod'

    Complain about this comment

  • 33. At 12:42pm on 20 Nov 2010, Gypsycruiser wrote:

    The spacecraft and the robots will be a unitry design,the robots being an extension of the craft or module.Humans will be accommodated but will play a very different role and very quickly the cyberbrain will be needed to make critical decisions.The partnership between humans and the cyberbrains will be the future.Scientists have furthered their interests by appeasing their political masters but are essentially apolitical but have elaborate ways of extracting funding.Cyberbrains will design and configure the best robots for future missions.Earths economy and trading standards will be greatly altered when for example precious minerals are returned in quantities that change our political views.The ExcoEconomics will be earths trading partner untill a group of humans leave the earth never to return.We are close to fusion reactors providing unlimited energy,capitalism as we know it has failed,the human race will be even more divided because religion will cause further wars.Religions stop and end when we leave earth....untill new revelations?

    Complain about this comment

View these comments in RSS


Sign in

BBC navigation

BBC © 2014 The BBC is not responsible for the content of external sites. Read more.

This page is best viewed in an up-to-date web browser with style sheets (CSS) enabled. While you will be able to view the content of this page in your current browser, you will not be able to get the full visual experience. Please consider upgrading your browser software or enabling style sheets (CSS) if you are able to do so.