Video: JSC's Project M

Robots like these would be great for helping to set up a lunar base before humans arrive for permanent occupation. And if humans were in a space station in Martian orbit, they could help to set up a base on the Martian surface.

This is a good idea, IMO. We need to start deploying real time teleoperated robots.

Marcel F. Williams

That is the exact sort of flagship mission they need to start under this budget. Also, it is the perfect way to do cross-promotion with Avatar.

Good: Probably worth it for the promotional value alone

Bad: People will ask what it is for, and there better be a good answer other than promotion

Ugly: Power source

RC: "Ugly: Power source"

I can think of a several male/female power receptical arrangements that would be sure to give everyone a chuckle!

This is completely in line with Flex Path. For the Moon, you can get by with direct ops from Earth, unless you're at a lunar pole.

Not to be too nit-picky, but 2-leg ambulation may be a bit sporty to do in a 1000-day timeframe. I don't doubt that the upper extremities could be done, as per Robonaut.

Some folks at NASA Glenn recently looked at a Robonaut based concept for Mars using 4 "whegs" per vehicle. These would be directly controlled from an crewed spacecraft orbiting the planet. Locomotive stability was easy.

Hopefully this representation isn't too literal. Bipedal locomotion? Ostensibly this is to save weight over a wheeled base, but it introduces major risk.

The question people are probably going to ask is "can we do this?" when it ought to be "should we do this?" As RC posed it, there had better be a real reason behind the effort other than "because it's cool". I've heard projected budgets of $500 million, which sounds like a lot of money for a PR stunt.

Looks like a cybernetic life-form node will reach the Moon first.

Why does it have to look like a human? Shouldn't the robot's design be optimized to fit the mission?

If I was going to do a 1000 day mission to the moon, I'd do it via JPL or GSFC or LaRc...and I'd land a rover with some science instruments. What expertise does JSC have in such unmanned missions?

Don't get me wrong: Cool Robonaut..and what would be the point other than a PR stunt? Need a good answer to that.. I'd demo the Robonaut via ISS to ensure it works as advertised before putting it in a 1/6 g gravity well.

That would be an insanely cool, captivating mission. Definitely a good PR move.

I would assume that they would or could incorporate various instruments into the robots design. Instead of a hand maybe have a drill bit and microscope or something like that. i.e. the borg. Or even interchangeable parts.

Lastly, I would this would be the perfect type of mission to demonstrate and develop a technology. The human body form is obviously highly versatile. As others have pointed evolutions of that same robot could be used in a wide variety of future missions.

I wonder what the prediction was for the Hubble Repair robot and how this compares to that robot. If we're to learn something from that, I would think it would be robotics take longer than you think.

This would be good to do, but realize that those giving the prediction (1000 days) are the ones promoting the concept. Also, since this hasn't been done before, chances are it will take longer and cost more than predicted.

Land it at the LCROSS impact along with a shovel and some other tools.

It can do everything many, many field geologists with a hammer and a hand-lens can do, in principle. We're talking full tele-presence, with appropriate comms and ICT infrastructure. There would be many actual specialist geologists at the Earth-side of the system, rather than a more generalist astronaut on the surface. It's the game-changing technology we can believe in. It's one of those approaches that moves beyond the human vs robotics divide. We will still need to send humans into Deep Space, using emerging propulsion, life support, etc, but we can access the (nearby) surfaces in more intelligent, 21st century ways not possible in Apollo-style approaches.

My four and three year old children have already asked to watch the video a couple of times. The robot needs a toolbelt...

The 1000 days was mentioned but I never saw the cost listed. Another multi-billion dollar boondoggle?

Land it at the LCROSS impact along with a shovel and some other tools.

There is no mission for this, it's just a flag planting exercise. It's a cool idea though for sure, they just need to partner it with a real mission.

Sounds like you know something about it. So, why does it have to go to the moon? Less gravity? Seems like if this could be developed in 3 years for the moon that something similar could be developed and would be useful for here on Earth. In which case one should be able to find funding. So, is 1000 days realistic? Or is it just marketing bull?

This is the way to go guys. Build the base and *infrastructure* with robots. Then your space faring civilization will get off the ground quickly, and for much cheaper than other options.

We're only 30 years too late (see: "NASA's Advanced Automation for Space Missions").

PR stunt idea: have weekly get-togethers at the local science or astronomy museum where kids get to drive smaller robots around the surface of the moon, in real time.

RC, don't use the walking version, use a wheeled version, put an RTG or other nuclear power source in the base of it. A human needs 100 watts continuous, it's not that big of a deal. I don't think they should go with a humanoid because it has more complexity than a walking version.

sc220, I think 2 legs is fully possible. Robonaut is effectively complete, they have 3 years to put legs on it. Yeah, dooable. But practical, I debate that. Why send such a robot unless it's going to be able to effectively and practically build stuff?

wtbard, problem with Hubble repair robot is that you needed to build a fairing beforehand to accept the Hubble and attach on to it. Hubble wasn't designed to accommodate a robot repair servicer, but a moon base could be designed completely around limitations that robots would have.

Land it at the LCROSS impact along with a shovel and some other tools.

I see a lot of people saying this would be a waste of money, but isn't this precisely what a proof-of-concept mission should be? Send it to the moon to test teleoperation feasibility, then move on to Mars, asteroids, etc.

Yes, it may not be totally groundbreaking in terms of the science that is done, but I'd rather the testing be done close to home to get the kinks worked out before committing really serious money to sending something outside of the Earth system. Wouldn't it be humiliating to put one of these things on a manned Mars orbital mission, but have it experience major problems when it arrives dirt-side?

Even with the costs involved, some useful science can be done, and never underestimate the value that a good PR mission can generate.

What is the TRL (Technology Readiness Level) of this?

Has it made it past 3?

This project has been floating around for several months now. I thought it had been dropped, but then this story reminded me...do the math! This is a GOVERNMENT project. What happens 1000 days from this Wednesday, February 10, 2010? Give up? The next Presidential election. So greenlight M by then and boom, Obama has returned us to the surface of the Moon!

It will be interesting to perform telepresence from a bipedal platform even with the relatively brief three second com delay to and from the moon.

From what I can tell, Robonaut, which has been around in various forms for ten+ years, is just someone's pet project that is basically a solution in search of a problem. The result is that it has become an engineering flourish with absolutely no science. Project M is consistent with that model, because I see no science in the proposal.

I've even heard that someone's pushing to get Robonaut on ISS. Why? Because there's only a few more shuttle launches and that window of opportunity is closing fast. Forget about what purpose it would serve, just get it up there. Definitely some salesmanship going on there.

J Smith: As for bipedal telepresence, mobility would have to be autonomous. There's no way you could have the operator's sense of balance translate to the robot walking with a 3 second delay.

Well, here's an idea: Let's just fight and bicker over what NASA should do, waste money on starting and stopping huge programs and never accomplish anything of any consequence. Ya, that sounds better.

Boondoggle? The only real boondoggle is investing 5B into a program and then killing it with nothing to show.

This technology is what is coming next. We should embrace it, not fight it.

Hubble Robotic Servicing & Deorbit Mission was going to use SPDM (DEXTRE)...trust me you can use DEXTRE in any gravity without offloading!

I get Shades of Cxp from that video. I am all for covering the moon with robotic drones be it humanoid or something more realistic and practical. Looking at where Human space flight is heading or isn't heading. I say Ramp up the robotic missions to the moon. Why not have HD pictures of Ice instead of going over data that is potentially inconclusive and not %100 accurate( I'm gonna regret that last one I know it). No Human Space flight certifications necessary.According to the Ares team that can take 5 years :|. Now don't get me wrong l'm a outspoken proponent of "boots on the ground" science.Why not Flood the Solar system with Robotic craft then?I know we have been but not like we could.Unfortunately Leo still cost so much .. Damn gravity!

So spending $500 million on a robot to go to the moon in 1000 days and having no useful purpose other than to say "we did it" is a better use of funds?

It seems to me the common thread of concern here is that there is no obvious value to this kind of mission as it's being portrayed. Given the country's financial predicament, you're going to have to offer something more than sensationalism.

So far, Project M's beauty is only skin deep. Is that all that NASA/JSC has to offer? I don't want to believe that but it seems that's what's driving the decisions nowadays.

err...meant CANT use DEXTRE in any gravity

The anthropomorphic robot is nifty and looks great, but unless you have something following it around the lunar surface, you'll never see it do anything. So who cares if it looks like a person? The form of the thing should be based upon it's mission. It looks like Rob Ambrose trying to hoodwink some NASA execs with shiny objects before they realize that it is not a mission of substance but of flash that will not materialize without a robotic tender following the thing around the lunar surface. YOU WILL NEVER SEE IT.

Now a cool Project M would go to someplace geologically fascinating to scientists and the public. Let say you figure out how to land on a mountain and have a HDTV camera equipped bot repel down multiple surfaces with an array of instruments. Of course build to the needs of the mission, not designed to look like a person "because it is soooooo coooool!" give me a break.

But the public will mock this mission when it turns out that all it is a robot that unnecessarily looks like a person for absolutely NO GOOD REASON and they never get to actually see it walking on the moon. Ok maybe they'll get some great shots of it's back as it walks away and fails. A robot corpse on the moon, yea, that'll encourage the kids. Sheesh.

I do not think that this in principle is a bad idea (ie robot on moon in 1000 days), but I would go with a proven EXISTING design and send a bunch of them. Would you rather NOT see a humaniod robot on the moon or would you rather look through the eyes of 10 to 15 opportunity/spirit class rovers modified for lunar duty?

This is just a dumb stunt to produce a cool video tacked on the heels of an awesome movie and it is a direct attempt to manipulate public opinion by using that movie as a PR tool. Dr. Ambrose and who ever he allowed to do this in the Robotnaut's team name should be ashamed.

Looks good, but Robonaut is "less filling"

Form should follow function. Pixar does these kinds of productions better.

How does the cool robot get to the moon? Is an Atlas V big enough? If you need a heavy lift capability 1000 days won't cut it.

Why does it have to look like a human? Shouldn't the robot's design be optimized to fit the mission?

I think there are a few reasons to assume that a humanoid shape is at least a good general-purpose starting point for this type of mission:

- It means that the object they are controlling is close to what the people "driving" it have learned to control all their lives: a human body with very specific degrees of freedom and movement types (walking, crawling, kicking, pushing etc.)

- As a result of millions of years of evolution, the human form can probably be considered a "proven" design in many ways

That said, some of the details as to how this shape is rendered into a practical design are probably owed to PR. Carry a few cameras on a stick to provide an outside perspective most likely won't hurt either.

At least for me, this idea instantly struck a chord when first seeing it.

LOL, I guess no one knows what the TRL of this is, or they don't dare say. I assume JSC knows about this thread. How does the time delay between the earth and moon affect your telepresence technology? Easy enough to simulate, so I assume you have. JSC, make a statement or cite some papers which put meat on the bones of this technology. And, make the papers that you have retained copyright to available on your web site.

A quote from JSC's web site for Robonaut (the statement may be dated)

"Advanced control theory in the areas of grasping, force control, intelligent control, and shared control must be developed to the point where the control is suitable for critical applications to fully realize the capability of the Robonaut."

Aren't these the issues all robot developers are still having? So, how far along have you come? 1000 days? Really?

2nd you mean: 20-7-69...!