Man Versus Robot!

[Warren Platts]

In the Moon Versus Mars! thread, an objection to sending people to a permanently manned Moon Base is that robots can do practically anything a human can do, and cost less, because they somehow don't require an infrastructure, whereas humans do. Naturally, my own position is that there is no way robots by themselves--or even teleoperated robots by themselves--can accomplish anything beyond a mere proof-of-concept pilot project demonstrating the feasibility of some industrial process or other on a place like the Moon.

I am using the term "robot" in its loosest sense as denoting any powered machine capable of replacing human labor. In that sense, a bulldozer is a teleoperated robot where the operator just so happens to be riding on top of the robot.

As a concrete example, consider an industrial plant on the Moon capable of generating several hundred tons of LO2 per year to be used for refueling orbital fuel depots. Such a plant would require that several modules sent from Earth would have to be bolted together. Do you think that robots alone could bolt together the required modules, maintain the plant, and then deliver the produced LO2 to orbital fuel depots using SSTO's that are themselves maintained by robots?

That is the poll question, but the discussion need not be limited to LO2 plants. Can robots do lunar geology and prospecting as well as humans? If not, are humans worth the extra cost? Perhaps there are important things that need to get done that can't be accomplished by robots at any price. In that case, humans are infinitely cheaper.

But let's stick to the realm of what is possible within the next few decades. Historical evidence, analysis of working examples of industrial processes on Earth, and mathematical engineering data is preferred to evidence taken from pulp science fiction.

[Zvezdichko]

I expected such thread to popup.

Nobody doubts that the cheapest way to do science is with robots. Robots can do it cheaper, but can they do it better? I doubt it... yes, as IsaacKuo mentioned yesterday - human actions are limited by the spacesuit - oxygen supplies... etc. Whatever... look what happened with Lunokhod 2. Look what happened with Spirit rover. They got stuck. A human being can easily unstuck them. Phoenix Mars lander photographed some droplets on its legs. Are they liquid water drops? We don't know for sure... If only we had an astronaut there..

Some may think the Moon is close so we can have teleoperated robots from Earth (while we can't have teleoperated robot on Mars unless we have people in Mars orbit or on Phobos or somewhere else close to Mars). This, some say, makes human spaceflight to the Moon unnecessary. Right? WRONG!!! After all, Lunokhod 2 got stuck and controllers couldn't get him unstuck. There are a lot of complicated operations that still can't be done by robots.

[Jens]

Whatever... look what happened with Lunokhod 2. Look what happened with Spirit rover. They got stuck. A human being can easily unstuck them. Phoenix Mars lander photographed some droplets on its legs. Are they liquid water drops? We don't know for sure... If only we had an astronaut there.

There may be times when a human is better, but there are times when robots do better. Think of the crew of Soyuz 11. A robot would have survived the decompression. And think of the fact that the Mars rovers have been there for what, several years? No human can stay in that environment for such a long time without a lot of infrastructure. The advantage of robots isn't necessarily that they're better than humans, but they can be designed for a particular mission or environment.

[Zvezdichko]

Jens: You're right but the example with MERs is incorrect. Steve Squyres already said that all of the year-long work can be accomplished within one or two EVAs.

[djellison]

You bemoan a single robotic rover getting stuck. You claim it to be a justification for a multi tens of billions of dollars manned program to the Moon. For a FRACTION of those tens of billions you could send another rover. And another. And another. And guess what... another. Who cares if one gets stuck - send another. Still a tiny fraction of the cost of a manned mission to the moon.

We've not even scratched the surface of what robotics can do on the moon - we've barely tried. Stereo HDTV robots driving across the surface exploring with an immersive telepresence experience that is

If you've going to go to the expense of putting people into Deep Space - do it where it makes sense. Do it, as Squyres as said, where they add the most value. Yet again, you take someones quotes, and singularly fail to put them into the context in which they belong. That is intentionally misleading.

Man versus Robots?

Both. Send each to where they offer the most effective results.

[Garrison]

So your asking could yet to be designed robots operate a purely theoretical LO2 plant that can't even be shown to be practical let alone economic better than humans using non-existent launchers and yet to be developed life support technology. Well in that case sure the robots can do it; especially if the Moon pixies help them...

[Garrison]

I expected such thread to popup.

Nobody doubts that the cheapest way to do science is with robots. Robots can do it cheaper, but can they do it better? I doubt it... yes, as IsaacKuo mentioned yesterday - human actions are limited by the spacesuit - oxygen supplies... etc. Whatever... look what happened with Lunokhod 2. Look what happened with Spirit rover. They got stuck. A human being can easily unstuck them. Phoenix Mars lander photographed some droplets on its legs. Are they liquid water drops? We don't know for sure... If only we had an astronaut there..

Some may think the Moon is close so we can have teleoperated robots from Earth (while we can't have teleoperated robot on Mars unless we have people in Mars orbit or on Phobos or somewhere else close to Mars). This, some say, makes human spaceflight to the Moon unnecessary. Right? WRONG!!! After all, Lunokhod 2 got stuck and controllers couldn't get him unstuck. There are a lot of complicated operations that still can't be done by robots.

But is robots getting stuck or having limited capabilities an argument for sending humans or just building better robots? My personal view is that the latter is likely to be much cheaper than the former, especially as improvements in robotic technology are already being funded by the military and the commercial sector.

[Ilya]

I voted no. At least sometimes a human would be needed to clear jams, and otherwise fix unpredictable problems. Your mining stations won't need MANY humans; perhaps one is enough, but it will need at least one.

[Ilya]

But is robots getting stuck or having limited capabilities an argument for sending humans or just building better robots? My personal view is that the latter is likely to be much cheaper than the former, especially as improvements in robotic technology are already being funded by the military and the commercial sector.

At some point it may become cheaper to write off a stuck robot and to send a replacement, than to keep a live human on hand with attendant life-support system. In fact, assuming you have technology to do Moon mining at all (which is MUCH more difficult than it sounds, given that all existing mining expertise assumes existene of air), it may be cheaper from the start.

I have to rethink my "no" vote.

[Zvezdichko]

I think I'm starting to live in a Parkist society. Fermi paradox solved.

[Hop_David]

In the Moon Versus Mars! thread, an objection to sending people to a permanently manned Moon Base is that robots can do practically anything a human can do, and cost less, because they somehow don't require an infrastructure, whereas humans do.

Immersive telepresence and manipulators with more degrees of freedom were not possible due to constraints imposed by processor speed and power as well as bandwidth.

Increased processing power of the robot enables processing of sensory information at the site and thus cuts needed bandwidth.

I believe Moore's Law has a great potential for breaking down these barriers.

So it is my opinion that much more able telerobots are doable in the near future.

Nasa's telerobotic program was discontinued in 1997. Most of the literature I find on Google is mid to late nineties. Evidently folks sharing your views have decided telerobotic development is a waste of money.

Do telerobots with close to human dexterity and mobility exist now? Of course not. However you use the present tense in your poll. In that regard you are misrepresenting my position.

And of course mining robots would require an infrastructure (albeit much less than humans). To say they don't require an infrastructure also misrepresents my position.

[Hop_David]

I voted no. At least sometimes a human would be needed to clear jams, and otherwise fix unpredictable problems.

With telerobots, humans would have a telepresence. Solutions to unforeseen problems would come from humans, not a set of preprogrammed routines.

[Hop_David]

Naturally, my own position is that there is no way robots by themselves--or even teleoperated robots by themselves--can accomplish anything beyond a mere proof-of-concept pilot project demonstrating the feasibility of some industrial process or other on a place like the Moon.

Telerobots are already used in industry. Offshore drilling platforms use ROVs to work on places humans can't access. Rio Tinto is seen as a pioneer in using telerobots in mining.

These aren't lunar environments. But this shows their utility in hostile environments.

[TampaDude]

Telerobots are already used in industry. Offshore drilling platforms use ROVs to work on places humans can't access. Rio Tinto is seen as a pioneer in using telerobots in mining.

These aren't lunar environments. But this shows their utility in hostile environments.

I would posit that 500 or 1000 meters below the ocean surface on Earth is at least as hostile an environment as the lunar surface.

[Bluevision]

I voted "Yes," but that's just that they CAN do it. I believe robots can do it, and they probably would be able to do it for significantly less than humans, but I don't understand why you'd cut corners like that. Avoid sending people to one place which is arbitrarily a middle point, so you can send people to another, more distant place? In that sense, robot's don't do it for me.

But I believe that robots can do what you described.

[Hop_David]

I voted "Yes," but that's just that they CAN do it. I believe robots can do it, and they probably would be able to do it for significantly less than humans, but I don't understand why you'd cut corners like that. Avoid sending people to one place which is arbitrarily a middle point, so you can send people to another, more distant place? In that sense, robot's don't do it for me.

But I believe that robots can do what you described.

If robots could establish and use such infrastructure, establishing an extended human presence on the moon would become vastly less difficult.

[Warren Platts]

Wow I must admit that I am a bit dismayed by the results of the poll. Probably my fault since the poll question itself is rather vague. I have no doubt that clever probe designers could build a lander with a robotic arm that could scoop up some soil, place that into an oven, and then generate a thermos bottle's worth of LO2. Such a probe would be an engineering marvel, no doubt. Such a result would be interesting and useful, but it would not in itself be game-changing. That's what we want: a game-changing industrial process that will revolutionize beyond LEO space travel.

The "hundreds of tons" figure was not chosen out of thin air. I refer the interested reader to "A Commercially Based Lunar Architecture" by Frank Zegler, Bernard F. Kutter, and Jon Barr (2009). Their Year 3 plan of operations calls for a propellent flux through an L2 fuel depot of 243 tons and another 485 tons through a LEO fuel depot. And this does not count the fuel used by the SSTO lunar shuttle craft. So "hundreds of tons" represents a bare minimum capability to be worth it. Thousands of tons would be better.

I figure the size and complexity of such a plant would be on the order of that of a small land-based oil rig. Land-based rigs are a better analog than off-shore rigs, since off-shore rigs have the luxury of being fully "ground integrated" more or less in one piece and are then towed to the location. Land -based rigs also must be portable, but they have to be broken down into "modules" that can fit on a semi-truck tractor-trailer. A working rig is a maze of bolts, pipes, power lines, wires and sensors.

According to Zegler et al., lunar landers launchable from today's generation of launch vehicles can land about 20 tons at a time on the Moon--coincidentally about the weight typically carried by semi-truck tractor trailer rigs. So clearly it will take several modules. So the question is: Who or what is going to bolt all those modules together, and make all the connections.

And here's another question: Why aren't there fully robotic oil rigs? The answer is: IT CANNOT BE DONE AT ANY PRICE. If oil execs figured they could save money by going fully robotic, they would do it, because payroll is by far the largest ongoing expense of an oil rig. Their budgets dwarf NASA's. So if the oil industry cannot afford to create fully robotic oil rigs, then how in the heck is NASA or whomever going to be able to afford to design and create fully robotic rocket fuel plants for use in Outer Space?

And what happens when something breaks? Doug Ellison's argument above that it's cheaper to send a replacement than to do a repair with humans only works with comparatively smaller, cheaper, less functional probes like Spirit and Oppy. Once robots get to the expense of a Hubble Space Telescope ($1.5 billion), it becomes cheaper to send humans to repair them. A fully functional LO2 plant would cost much more than the HST.

Remember, we're talking about what's possible within the next 20 to 40 years. Humans could be back on the Moon in 10 if we set our minds to that task. If we have to wait around decades and decades and decades for pulp science fiction novels to come true, then Man wins, and Robots lose!
_________________________________
"The principal reason for going to the Moon: to use its resources to create new space faring capability."--Paul Spudis

[Warren Platts]

I think I'm starting to live in a Parkist society. Fermi paradox solved.

No kidding....

[Hop_David]

And here's another question: Why aren't there fully robotic oil rigs? The answer is: IT CANNOT BE DONE AT ANY PRICE.

You state this as a fact. Why is it impossible to make teleoperated robots having dexterity and mobility comparable to a human?

If oil execs figured they could save money by going fully robotic, they would do it, because payroll is by far the largest ongoing expense of an oil rig.

So you believe teleoperated robots would eliminate payroll expense? No. You evidently don't know what teleoperated robots are. You must pay an employee to operate a telerobot.

Also, as I have mentioned, immersive telepresence and manipulators with many degrees of freedom required computer processing power that wasn't available. Moore's Law is changing that.

Their budgets dwarf NASA's. So if the oil industry cannot afford to create fully robotic oil rigs, then how in the heck is NASA or whomever going to be able to afford to design and create fully robotic rocket fuel plants for use in Outer Space?

And what happens when something breaks? Doug Ellison's argument above that it's cheaper to send a replacement than to do a repair with humans only works with comparatively smaller, cheaper, less functional probes like Spirit and Oppy. Once robots get to the expense of a Hubble Space Telescope ($1.5 billion), it becomes cheaper to send humans to repair them. A fully functional LO2 plant would cost much more than the HST.

The first successful uses of CCD arrays were in satellites where radiation destroyed film. In those days film worked better than CCDs for everyday terrestial uses.

But after we started making CCD arrays for satellites, they migrated to commercial cameras. Then very powerful CCD arrays became very inexpensive.

I don't think it's unrealistic to hope for a similar path for teleoperated robots. The first telerobots are being used in places where actual human presence is impossible or extremely difficult.

But as they evolve, they will increase in utility and become used in environments where human presence is only moderately difficult. When market forces and economies of scale kick in, it might become possible for the average Joe to use a teleoperated robot to avoid commuting to work.

If this came to pass, telerobots needn't cost 1.5 billion apiece.

Humans could be back on the Moon in 10 if we set our minds to that task.

Quite true. Unfortunately that's a big IF. A Mt. Everest IF. You're indulging in wishful thinking here.

Given lack of public interest, telerobots seem a more plausible path to using extraterrestial in situ resources.

And if that comes to pass, an extended human presence becomes much more doable.

[Warren Platts]

You state this as a fact. Why is it impossible to make teleoperated robots having dexterity and mobility comparable to a human?

I saw Blade Runner too. It was a great movie. The point of this thread, though, is whether robots can compete with humans--if not now, then at least within a few decades. If robots are to be used as a justification for postponing plans to send humans to the Moon and other destinations then it would be nice if they could actually get out there and do it for us while we're waiting around. If we have to wait around for such robots to finally get invented, and then wait around some more while they explore for us, then that's a double wait-around, and hence unnacceptable.

So you believe teleoperated robots would eliminate payroll expense? No. You evidently don't know what teleoperated robots are. You must pay an employee to operate a telerobot.

Good point! Maybe it would be better to skip the layer of complexity and cost that the telerobot represents!

Also, as I have mentioned, immersive telepresence and manipulators with many degrees of freedom required computer processing power that wasn't available. Moore's Law is changing that.

Such computing power is still not available, and Moore's Law may run into a brick road soon.

The first telerobots are being used in places where actual human presence is impossible or extremely difficult.

On the contrary, telerobots have been with us for a long time. A bulldozer is sort of a telerobot. But humanoid telerobots kind of defeat the purpose of most telerobots. Most robots do things individual humans can't do; why create a robot that is designed to do what an individual human can do?

But as they evolve, they will increase in utility and become used in environments where human presence is only moderately difficult. When market forces and economies of scale kick in, it might become possible for the average Joe to use a teleoperated robot to avoid commuting to work.

I can picture it now: some day nobody will have to actually go to work. They'll just climb into their virtual reality suits, wake up their particular robot at the job site, and then pound nails for 10 hours. I think I would rather just go and pound the nails myself.

If this came to pass, telerobots needn't cost 1.5 billion apiece.

I certainly hope not!

Quite true. Unfortunately that's [Humans could be back on the Moon in 10 if we set our minds to that task] a big IF. A Mt. Everest IF. You're indulging in wishful thinking here.

I don't think so. The "Moon Versus Mars" poll results here show that interested bautforum.com registered users prefer Moon First two to one over any other option. The "Let's-Use-Telerobots" option is a version of the Neither option. That is the option that lacks public interest.

Given lack of public interest, telerobots seem a more plausible path to using extraterrestial in situ resources.

And if that comes to pass, an extended human presence becomes much more doable.

I think you really misunderstand the American people. Americans have chosen to spend on the order of ~ $10 billion USD per year on human space flight year after year after year. So your point about public interest or disinterest is really beside the main point. The fact is that Americans are interested (some might call it disinterested I'll grant) in human space flight to the tune of $10 billion USD per year. That money has to be spent on something. Thus, to take that money and spend it on telerobotic space flight instead doesn't make sense to me. But that question is moot because telerobots aren't even up to the task. Not to mention that 2.5 second round trip the wireless signal has to make.
_________________________________
"The principal reason for going to the Moon: to use its resources to create new space faring capability."--Paul Spudis

[IsaacKuo]

Why aren't there fully robotic oil rigs? The answer is: IT CANNOT BE DONE AT ANY PRICE.

Actually, there are plenty of pumpjacks which pump away day in and day out with little ongoing human presence. You see them dotting the landscape in many oil rich areas.

If oil execs figured they could save money by going fully robotic, they would do it, because payroll is by far the largest ongoing expense of an oil rig.

This is why pumpjacks are essentially unmanned. It's cheaper that way. Sure, humans do the initial installation and perform occasional maintenance--because it's cheaper that way. But these humans don't need to travel to the Moon and back. They don't need to do their job wearing space suits. They don't need expensive life support supplies.

Anyway, the task of mining the surface of the Moon for resources is far simpler than deep drilling.

[Hernalt]

"Do telerobots with close to human dexterity and mobility exist now? Of course not."

For comparison:

Toyota Violin-playing Robot
http://www.youtube.com/watch?v=z9sirG0UEGg

Aldebaran Nao
http://www.youtube.com/watch?v=rSKRgasUEko

Big Dog
http://www.youtube.com/watch?v=azoWDlZGImU

I'm enjoying this spirited discussion of this specific problem.

In response to the logic that X oil jobs |cannot| be done by robots because western market forces |would have already| achieved a cheaper robotic workforce, I would have to politely ask if those humans on the oil rigs are paid, in equivalent terms of comprehensive and redundant life-sustaining infrastructure, as much as astronauts. As a thought experiment, evacuate the oil rig of atmosphere, liquid water, thermal and radiation insulation, but leave the precious commodity exactly where it is. Somewhere on the supply/demand curve is a point of economic return that justifies development of a teleoperated replacement of human labor.

The question - and here I want to see more sparks fly - is what model, frequency, or logistics of necessary human servicing missions could be integrated into a scaled-up lunar presense of rigourously tested telerobots? High fault tolerance should mean fewer servicing missions. Hubble servicing missions were as-needed for specific failures and upgrades. Could ground-based robots be made with similar schedules or expectations of mean time to failure?

As for the problem-solving and adaptability equation for which humans are the only ostensible solution, improved environment immersion and improved telerobot dexterity should positively impact the schedules for how frequently humans must be sent to the moon. The positive steps taken in robotics in the above links in no way supports the idea that some hypothetical all-robot infrastructure on the moon could maintain business as usual indefinitely. All automation needs servicing, usually periodic, sometimes ad hoc.

Another aspect is the scaling factor. Beneath some initial investment or quantity of telerobots, it does not make best sense to support the overhead of a permanent human base. But above that minimum scale...

I hope to see more discussion on where the human-stops / robot-starts line is drawn.

[Warren Platts]

"Do telerobots with close to human dexterity and mobility exist now? Of course not."

For comparison:

Toyota Violin-playing Robot
http://www.youtube.com/watch?v=z9sirG0UEGg

Aldebaran Nao
http://www.youtube.com/watch?v=rSKRgasUEko

Big Dog
http://www.youtube.com/watch?v=azoWDlZGImU

Nice! But I don't Yoyoma is about to lose his job!

I'm enjoying this spirited discussion of this specific problem.

In response to the logic that X oil jobs |cannot| be done by robots because western market forces |would have already| achieved a cheaper robotic workforce, I would have to politely ask if those humans on the oil rigs are paid, in equivalent terms of comprehensive and redundant life-sustaining infrastructure, as much as astronauts. As a thought experiment, evacuate the oil rig of atmosphere, liquid water, thermal and radiation insulation, but leave the precious commodity exactly where it is. Somewhere on the supply/demand curve is a point of economic return that justifies development of a teleoperated replacement of human labor.

There might also a point on that demand curve that justifies buying space suits. Also you have the time delay factor to think about. How can you thread a needle with a 2.5 second time delay. It would drive me crazy anyway. And if you're going to send people to an L1 space station where they can virtually operate robots on the surface, then why not go the extra mile and land on the surface yourself?

The question - and here I want to see more sparks fly - is what model, frequency, or logistics of necessary human servicing missions could be integrated into a scaled-up lunar presense of rigourously tested telerobots? High fault tolerance should mean fewer servicing missions. Hubble servicing missions were as-needed for specific failures and upgrades. Could ground-based robots be made with similar schedules or expectations of mean time to failure?

Why not?

As for the problem-solving and adaptability equation for which humans are the only ostensible solution, improved environment immersion and improved telerobot dexterity should positively impact the schedules for how frequently humans must be sent to the moon. The positive steps taken in robotics in the above links in no way supports the idea that some hypothetical all-robot infrastructure on the moon could maintain business as usual indefinitely. All automation needs servicing, usually periodic, sometimes ad hoc.

Another aspect is the scaling factor. Beneath some initial investment or quantity of telerobots, it does not make best sense to support the overhead of a permanent human base. But above that minimum scale...

I would draw that minimum scale pretty low, because the overhead cost of keeping people on the Moon once you get them there is quite low compared to the cost of getting them there and back. In other words, if you are going to go through the time, trouble and cost of sending people to the Moon, you want to get as much work out of them as you can. That means long tours of duty, and so if you are going to have people there at all, the man-hours per dollar invested go way down if you have a permanent base.

[Warren Platts]

Actually, there are plenty of pumpjacks which pump away day in and day out with little ongoing human presence. You see them dotting the landscape in many oil rich areas. This is why pumpjacks are essentially unmanned. It's cheaper that way. Sure, humans do the initial installation and perform occasional maintenance--because it's cheaper that way.

Pump jacks only appear unmanned. There is a guy in a pickup truck who checks up on each one about once a day.

[djellison]

Why aren't there fully robotic oil rigs? The answer is: IT CANNOT BE DONE AT ANY PRICE.

Please provide ANY evidence that a robotic oil rig can not be done AT ANY PRICE. Any evidence at all.

Please tell me you're not saying that. That's as bad as me claiming that ISRU on the moon will never work because they don't do ISRU at the Antarctic base ( an argument I have made, then explained immediately why it's not even slightly appropriate)

You're forgetting something very very fundamental. It costs maybe $1000 to fly an engineer out to an oil rig. It will cost probably a million times, or more, than that, to send an engineer out to the moon.

What you SHOULD have said is that it's currently cheaper to have manned oil rigs. That is almost certainly true given the fact that it's very very cheap to fly people and supplies out to an oil rig.

Seriously - how can you even start making an argument like that - how utterly detached from any realistic grasp of engineering, budgets, hell, common sense- must you be to try and make an argument based on the lack of robotic oil rigs. E

Once robots get to the expense of a Hubble Space Telescope ($1.5 billion), it becomes cheaper to send humans to repair them.

Actually - it's widely regarded that it would have been cheaper to replace Hubble than conduct the repair missions. Google 'Hubble Origins Probe'. Plus guess what - Chandra. XMM Newton. Spitzer. Galex. Swift. Integral. WISE. COROT, Kepler, Herschel, Planck. No human has ever been to repair any of them. Nor will they - it would be far far cheaper to simply launch a replacement probe, a better probe, a more accurate, sensitive, stable probe.

A fully functional LO2 plant would cost much more than the HST.

Yet you repeatedly tell us it would be assembled from multiple moderate sized modules. If required, could they not be replaced? What type of failure are you expecting that could not be fixed by the sort of dextrous telepresence robotics cited in this very thread?

Remember, we're talking about what's possible within the next 20 to 40 years. Humans could be back on the Moon in 10 if we set our minds to that task.

Minds don't pay for things. Cheques do. Very very big cheques in this case, and no one's going to write one big enough to do that.

[Jens]

I think I'm starting to live in a Parkist society. Fermi paradox solved.

You must have meant that facetiously, but who knows, you might be closer to the truth than you intended. That is indeed a possible solution of the Fermi paradox. Maybe all the other civilizations in the galaxy have decided it's too difficult to send "humans" or whatever they are into deep space.

[01101001]

Wow I must admit that I am a bit dismayed by the results of the poll. Probably my fault since the poll question itself is rather vague.

Poll-taker's lament: I should have asked a different question to get the result I wanted.

[IsaacKuo]

You must have meant that facetiously, but who knows, you might be closer to the truth than you intended. That is indeed a possible solution of the Fermi paradox.

Not exactly...

Maybe all the other civilizations in the galaxy have decided it's too difficult to send "humans" or whatever they are into deep space.

The Fermi paradox actually doesn't assume space travel, and the most famous way to analyze the Fermi paradox--the Drake equation--doesn't even factor in the possibility. It's not clear exactly how one should interpret the last factor in the Drake equation, but the most straightforward interpretation is that all technological civilizations destroy themselves and the biosphere they rode in on before getting much further than radio transmissions.

Anyway, getting back on topic...if we are to interpret the Fermi paradox in a non-Drake equation fashion, the most popular question is to wonder why they aren't already here. In which case, it doesn't really matter whether "they" are "humans" or "robots". Indeed, self propogating robots would seem to be a lot better at expanding than biological creatures.

[IsaacKuo]

Pump jacks only appear unmanned. There is a guy in a pickup truck who checks up on each one about once a day.

In that sense, unmanned Predator drones only "appear" unmanned.

Anyway, they pay a guy in a pickup truck to check on the pumpjacks because it's cheap enough to do this. It actually makes more economic sense to not design pumpjacks to be excessively reliable. Besides the extra costs to make it more robust, you risk losing the maintenance worker's expertise. A field of pumps where there's a breakdown on a regular basis keeps the worker's skills up to speed. But if it cost billions of dollars to have the maintenance worker around? That changes the economic equation.

[hypergreatthing]

ohh a new robot thread. My type of argument.

So... can someone explain to me the term "telerobot"? I think that defeats the purpose of being called a robot which has the concept of automation around it. Even being programmed to do certain functions something can be called a robot or robotic. The idea that a human is controlling this tool's every function seems to be out of place with the concept of a robot. I don't call an RC racing car a robot, and that's what essentially this "telerobot" would be, is a tool that is operated remotely by a human doing human activities. I can't accept that my car is a robot since i am driving it and it is providing transportation and replacing the human equivalent to walking or running.

So if your asking if tools which are controlled by humans can operate a factory, well probably. Would it be easier with humans being there on site? Probably.