Will Robonaut 2 usher in a new age of man-machine space exploration? No.
Sorry to kill your buzz, NASA and GM, but your jointly developed humanoid robot, the Robonaut 2, is way over hyped. The 300 pound bot, which consists of a torso, two arms, and shiny gladiator looking head, is currently in the belly of the space shuttle Discovery set to arrive at the International Space Station. It’s “the first humanoid robot in space”! …Whoop-di-freakin’-doo. NASA has put dozens of robots into space, and had them perform amazingly complex missions both remotely operated and autonomously. Why should this one get all the press simply because its shape is humanoid? And the things that are being discussed – Robonaut helping astronauts perform repairs in space, being paired with humans for long range space missions, being set loose on Mars – these things are years (if not decades) away from being achieved. Robonaut is being sent into space simply so it can be tested to make sure it works in the new environment. All these other bits of hype are the worst kinds of speculation: they get our hopes up so high we won’t appreciate the real achievements we’ll end up seeing. Seriously, I’m not one to bad mouth a good piece of technology (and Robonaut 2 is a fine piece of electronic wizardry), but the hype surrounding this robot is some of the most ridiculously misleading and misguided PR claptrap out there. Check out the Robonaut videos, and much more of my ranting, after the break.
Robonaut 2 should be properly viewed as a research robot. It’s a device that can help engineers and scientists understand how humanoid machines function. With five fingered hands, a sensor-packed head, and relatively strong arms (capable of lifting 20 pounds at full extension), the robot is a good model for how the human form can be used as a model for building dexterous machines. Robonaut 2 is not, I repeat, NOT a functioning tool for space exploration. To get to that level could take decades.
That’s not the expectation you would get if you saw much of the media coverage about Robonaut 2′s trip into space from the past year. Here’s a clip from ABC that’s pretty typical:
Project M, which is discussed in the second half of the video, is a yet-to-be-funded endeavor aimed at putting a humanoid robot on the moon by 2013. Why a humanoid robot when a rover or exploration drone might be much more useful, not to mention cheaper? Ostensibly because humanoid robotics are such an important avenue of research, but I’ll tell you the real reason: humanoid robots are more exciting! Project M is a PR stunt. A well intentioned one, probably aimed at generating some much needed interest in space exploration and robotics, but a PR stunt all the same. And Project M isn’t the only PR stunt that involves the Robonaut. NASA has the robot tweeting, making Facebook status updates, and generally anthropomorphizing the heck out of it. GM has been hyping the bot as well, going so far as to buy some air time before the Superbowl to show-off Robonaut’s shiny golden frame. GM, the same automotive company that has struggled to stay solvent, is buying some of the most expensive TV coverage possible just to generate some buzz around all the great technology it’s helping to develop. What a stupid investment of time and money.
Yet some people are buying the hype! I really have to pick on Engadget readers who voted the Robonaut 2 number one in their poll for Best Robot of 2010. Wow. It didn’t even make it on our list. The thing you won’t find in any of Engadget’s prolific coverage of Robonaut is a rational explanation as to why NASA (and GM) are spending so much money selling everyone on the idea of humanoid robots in space.
You know who likes humanoid robots? Humans. We like the human form, we identify with it. Humanoid robots are going to be an important part of getting people to accept automated machines in sensitive industries such as elder care. Japan is working over time developing life like humanoids that people will feel warm and caring towards. Grandma won’t like her bedpan being changed by a robot, but if it looks like a friendly woman maybe she’ll complain a little less. And if you give a robotic baby she can hold and use as a phone (seriously), well things just got that much better. Human-like robots make sense in human environments.
You know which environment has, like, zero humans? Outer-freakin’-Space!
Time and again, though, NASA has painted the picture that Robonaut’s successors will be used in space travel. They have tons of animated footage showing a golden robot exploring the Moon, Mars, and Earth’s orbit – often accompanied by humans it is assisting. Here’s a video that puts all these clips together:
Gimme a break, NASA! Are you seriously going to tell us that the best robot for a job in space is a shiny metal humanoid that can hand briefcases to human astronauts? Not to mention the fact that you keep showing Robonaut walking around on legs. Bipedal motion, especially in changing gravities, is incredibly difficult to master. The Robonaut project hasn’t even started developing legs yet. You literally have no legs to stand on, NASA!
I find this so frustrating because NASA has created some amazing robots in the past. Robots that it has already launched into space, and already landed on Mars. There are, in fact, amazing robots on the International Space Station waiting for Robonoaut 2′s arrival. The Canadarm2 is a long telescoping robotic arm that helped build the freakin’ ISS in the first place! Dextre is a super articulate manipulator whose arms and hands are capable of some amazing tasks, equal to and exceeding human levels in some cases. And those hands have built-in tools! Why design a humanoid robot that can pick up a tool when you can design a better robot that has the tool already inside of it?
Humanoid robots would only make sense if you anticipate having humans involved in every step of future space exploration…and honestly, I just don’t see that happening. Humans aren’t designed for deep space, but robots can be. We shouldn’t handicap ourselves by thinking that robots need to look human in order to succeed in space.
As I mentioned in the beginning, the serious damage from all this humanoid in space hype, however, is that it will obscure the real benefits that this ISS mission for Robonaut 2 could achieve. NASA has us expecting robots and humans acting together in some sort of Star Wars style mission. Robonaut 2 is not R2D2. You know what Robonaut 2 will be doing while at the ISS? Here’s a video to give you some idea:
That’s right: tedious, tedious testing of maneuvers. That’s pretty much all Robonaut 2 is slated to do in space. NASA has to learn how the robot reacts to the microgravity, the radiation, and the electromagnetic interference. Other robots, like Dextre, have had a long history of malfunctioning in the dangerous environment. It can take years or troubleshooting to get a space robot fully functional. But read this quote from NASA:
“R2 will be confined to operations in the station’s Destiny laboratory. However, future enhancements and modifications may allow it to move more freely around the station’s interior or outside the complex.”
—Description of International Space Station mission on Robonaut website (2011)
That first sentence is all we really need. Yet they can’t seem to avoid the hype-machine. They always have to push it.
Let me give you a much more reasonable expectation for what Robonaut 2′s trip to the ISS will be all about: NASA will perform some boring and routine tests of the robot that will greatly resemble work that has already been done on the ground. After (what will probably be) months of problem solving, Robonaut 2 will start to be able to perform slightly more complex maneuvers that have nothing to do with the safety or function of the ISS. NASA might learn a thing or two about robots operating in space. Maybe we can use some of this research back on Earth if it can be generalized to systems outside of NASA’s hardware/software. …That is all.
This photo gets it right:
Robonaut 2 in storage for transit.
This video gets it wrong:
Look, I think Robonaut 2 is a great piece of technology. Humanoid manipulation isn’t exactly easy, and this bot does it fairly well. It’s not the best, but it’s pretty good. The harsh environment and rigorous testing aboard the ISS may provide some great insight into humanoid robotics in general. I’m not denying any of that. These positives are just getting oversold to the public. …and maybe I’m even a little upset because the hype might actually work. NASA and GM aren’t in the best financial situations, but they are getting some major attention from Robonaut 2 and its “potential” uses in space. Even if the robot isn’t the best investment in the world, the PR surrounding it could be. Does that make all this hype even more sad…or ingenious? You decide. I’m going to go buy some more exclamation points and inflammatory statements at the hackneyed writer’s store. Somehow it seems I’ve run out.
[image credits: NASA]
[video credits: ABC Campus News, NASA, lettiecat]
[sources: NASA]
Comments
I think this article describes the subject clearly and from a subjective point of view.
That is,
1) Robonaut is an impressive piece of technology
2) Unfortunately many seem to think that this humanoid is far more advanced than it really is, thus creating a larger hype than it’s worth.
Though as previously commented, “waste of time and money” seems a bit harsh.
Nevertheless, it’s still an interesting topic, so thank you.
Amazing stuff !
But we need a spaceship.
We need a spaceship.
Need a spaceship.
A spaceship.
Spaceship.
Spaceship for what? If we’re going to send robots throughout the solar system and ultimately to the stars, we may as well just build bigger telescopes. Much cheaper than sending anything at all. The alternative would be to send a human, but here at the Singularity Hub I guess we are mostly self-haters and would rather our race went extinct in exchange for a machine ruled universe.
As a long time Robonaut fan and acquaintance of some of the staff I am delighted to see all the “hype”. NASA really needs something to excite the public and maybe this can be it. And, as mentioned in several posts this may be just the thing to interest younger kids in robotics.
Just because this robot may not be the ideal configuration to explore the moon or work on the ISS doesn’t mean that its a PR stunt. Since its a huminoid form it should be adaptable to a variety of environments and configurations, including manufacturing since it was funded by GM. I think the spinoff factor has to be heavily weighted here before one labels it completely PR.
Aaron Saenz thinks NASA Robot is a waste of time, and he’s entitled to his opinion. Equally I feel his ranting in this article were a waste of my time. Technology is more than being the latest and greatest, it’s about inspiring and getting some young child involved in a career of their dreams. Wonder how many bad books Aaron read before he became a “writer”
Your thoughts are so very limited in your conceptualization of both the mission and roles for RN2 and its amazing that your naive intuitions even make it to the blog. EVERY thing NASA does is an experiment by nature. You also know the tride and true rule #1 of space exploration… no bucks… no buck rogers! I don’t think the hype is sad, you on the other hand should go work for the peace corps with your happy demeanor… perhaps in a middle eastern area full of the most tolerant of mankind!
I don’t believe that your rant is entirely called for. While it is, some way, a “PR stunt”, a humanoid robot would be very useful operating in EVAs where one misstep could result in an astronaut’s life being lost. R2′s descendants could make the old sci-fi scenario of the astronaut endangering his/her life doing vital repairs on EVA, a thing of the past.
Further, I think you underestimate the versatility of the human hand. If your robot has hard-built in end-effectors, what happens when it encounters a situation for which no one had prepared the ‘proper built-in tools’ for? In the uncertain environment of space, the ability to use the same tools that humans use may well be invaluable.
I need to agree with Aaron. NASA’s public relations videos are dreadfully dry and boring. The stuff online that makes people’s mouths water was all produced by GM and the TV networks. R2 is only the first gen humanoid in space. and if a humanoid makes it further than the ISS, it’s going to be attached to an arm attached to a ship or probe, or a wheeled/tracked platform. Bipedal motion on another planet is tricky even for humans. I can’t remember the mission or the astronaut involved, but someone tripped over a lunar equipment setup, and ruined the test, if not the entire experiment. If there is a bipedal R2 on the moon, and it falls backwards, in it’s current configuration, the entire project will grind to a halt while NASA tries to get it to roll over to pick itself up. Hell, they can’t even rescue Spirit from it’s sandtrap on Mars! No, a humanoid isn’t going to be commonplace on space missions anytime soon.
I can’t pay no doctor billz, but robot is on the moon.
http://www.youtube.com/watch?v=PtBy_ppG4hY
Its obvious the author doesn’t get it. The way to get funds to develop new technology is to make it appealing to the general masses (read taxpayers) that don’t understand the complex stuff. R2 does that brilliantly and that alone is reason enough to continue developing humanoid robots.
Does it really make any sense at all to give money to GM to develop a humanoid robot? GM is a dinosaur and is definitely not a software company. When it comes down to it, humanoid robotics really is a software problem.
this kind of robots will be very useful to explore Mars from a manned orbital space station on Phobos
I don’t think the author is trying to deny the usefulness of robots. He is just questioning that of a humanoid robot. I agree that 2 arms and delicate hands should be very useful in some cases. You would want camera on the robot so why not add the head, too. But the 2 legs? Do the humen astronauts use their owns that much? There are probably a lot of better and easier alternatives.
Whoever thinks this project was a waste of money and is completely useless, and is one to blame for America’s lack of technological innovation and creativeness in the global environment.
More than likely, you used technology developed for and from past space programs to help you electronically share your worthless and misinformed opinion with the rest of the world.
U mad?
This isn’t Huffpo. People that visit this site are most likely going to be NASA and humanoid robot fanboys.
You go on believing that the US lacks technological innovation and creativeness.
There have been humanoids built which were of not much more than entertainment value, but I don’t think this is one of them. It’s going to be worthwhile to find out whether teleoperated humanoids can do useful work in space, and if they can then there are many applications. The only thing which is certain is that astronaut based maintenance and construction is not the way forward.
When dealing with space hardware the high costs of getting it above the Earth’s atmosphere mean that conservative assumptions need to be made. In a conservative scenario we can assume that there are no significant advances in AI in the near future, and that robotic manipulation tasks are typically going to require teleoperation. The current methods of assembling and maintaining large scale structures in space using astronauts isn’t scalable for a variety of reasons, but it would be using telerobots operated by a ground based workforce, provided that the distances and time delays are not too great.
The anthropomorphic form just makes human teleoperation as easy as possible. You can have telerobots stationed permanently in orbit or on the moon, and they can work on construction tasks in shifts 24/7 in a way which requires little specialised training for the operators (i.e. you can hire construction workers on regular wages, rather than astronaut wages).
Having sophisticated and scalable construction and maintenance ability is an essential prerequisite to being able to do other things like asteroid mining, creating large moon bases or non-trivial deep space missions involving humans.
Probably the most impressive thing the Robonaut could do would be to use tools like drills, saws and spanners on a test rig set up inside the ISS, with the operator either based inside the space station or on Earth. If it can do those things competently in a microgravity environment and in a way which is not too taxing for the operator then there is a big future for this sort of robotics. If not, then we just have to wait until AI is good enough to do the same jobs – which at the present rate could be quite some time.
Wow, haters gonna hate.
Also, “The Robonaut project hasn’t even started developing legs yet.” is entirely untrue. Sorry.
What a stupid article… can I please have the 10 minutes back I wasted reading this drivel… You seem to have a hard time grasping the concept that generating PR generates money which NASA and GM are badly in need of. You also seem to miss the fact that this robot will be operating in human environments which necessitates a humanoid form, it also will be going on space missions and can teach us the human constraints of doing so by testing the facilities before we get there.
Next time waste your energy writing an article that really matters.
We’re just telling it like it is. This robot is a waste of time and money. Its only value to NASA is for PR – is that really enough reason to pursue the project? I know how PR works, but do you know how it works? NASA is a government organization – PR like this is very unlikely to change its budget – political forces, not PR are deciding NASA’s budget, and in today’s climate of super tight budgets, there is NO extra money for NASA, even with some humanoid robot PR stunt.
i believe you are telling it like it is… but anyway, you know, honestly i don’t think that robots worth such a great investigations. and i suppose that it is a huge amount of money that scientists spend on this.. considering all the facts of the real world that we all are live in – space robots is not what we really need..
Jeez Keith, I usually love your posts, but this article gives the impression you’re missing half your brain or something! First, chill my friend! A humanoid robot would be very useful in human environments, which includes human-ferrying spacecraft because we can relate to them, and they could perform a variety of tasks. Re PR: PR DEFINITELY affects budgets! Who do you think those politicians have to answer to? Their constituents! And if those constituents are clamoring for humanoid robots, they’re more likely to be funded.
We still love you, Keith. Everybody has a bad day.
The simple fact is that this robot currently just does not DESERVE a ride on the space shuttle.
Would you mind elaborating on precisely what you think the mechanism is whereby NASA, a government agency, generates money from this?
The article got it right. NASA, the space industry, and society in general needs to get serious and go back to the scientific, engineering and practical imperatives of space exploration and exploitation.