Human versus robotic spaceflight

[Jimmy Snyder]

I think you completely missed his point. He doesn't want to pay to send someone into space just so an astronaut can enjoy the view, when we can send a robot to do it at half the cost.

Half?

 

[mgb_phys]

It's estimated that Hubble cost 3x as much to build due to the shuttle launch - part of this was the cost of storage after Challenger, but most was the much tighter specifications for anything that is going into space with a human.

Ironically the first plans for the space telescope in the 70s pictured it as a manned observatory, like skylab, - with astronauts developing photographic plates.

 

[Contrapositive]

Half?

An arbitrary example. The point is if it is cheaper to send a robot, why sent a person?

 

[D H]

There are many things humans simply cannot do in space. We cannot not fly humans into space to operate a camera or communications relay 24/7 for several years. We cannot fly humans to Mercury or Pluto, and probably won't be able to do so for centuries. The extremely mundane tasks and the humanly unreachable destinations are jobs for automated vehicles. There is a place for and a need for robotic space flight.

There are similarly things robots cannot do in space. They cannot make snap judgements. Robots, in fact, are incredibly stupid and painstakingly slow. Early in the rover missions to Mars there was much excitement over the analysis of samples drilled from a rock by one of the rovers. The sampling took days; the analysis took even longer. A scientist-astronaut could have deduced the nature of the same rock in a glance.

All of you unmanned space enthusiasts have missed my main point. One of the biggest things human space flight does for unmanned space activities is provide a rationale for the very existence of the unmanned space activities. Unmanned space enthusiasts hypothesize that ending human space flight would free up vast quantities of money to spend on unmanned space flight.

Think of the end of Apollo and BNSC's decision to preclude frogs in space as scientific experiments of this hypothesis. Did vast sums of money flow to JPL and GSFC post-Apollo or to BNSC post 1986? No. The exact opposite happened with NASA at least (I can't track historical BNSC budgets). The budget for NASA's unmanned exploration efforts fell with commensurate with the rest of NASA's budget. These "experiments" falsify the hypothesis.

Unmanned space activities are not cheap. To the contrary; they are very, very expensive. The Cassini mission, $3.27 billion. The Mars Exploration Rovers, $850 million. That represents funding for thousands of scientists for several years plus an untold number of graduate students. Cassini, Spirit, and Opportunity have yielded incredible results, but at an incredible cost. Will Congress fund future endeavors such as these without the added impetus of being a pathfinder for human activities? I doubt it.

 

[mgb_phys]

Of course the point of a Hubble/Cassini/etc is to fund hundreds of grad students/postdocs - the only place the money for a manned space mission goes is to Rockwell/Boeing/MortonThiokol.

Unfortunately you are right about the money following public/media interest - it's a similair problem in the military. Nobody wants unmanned drones/missiles because you aren't going to get elected based on having heroically flown a computer terminal in Gulf war III.

 

[russ_watters]

Pros of human spaceflight

• On-the-spot judgments, innovation, adapt to surroundings, react quickly to the unexpected, flexibility

• This was the subject of much discussion on page two, but it is a complete non sequitur. The 50% success rate of sending probes to Mars was cited as a reason why you should have humans. But with combined total mission price of under $1 billion, we could have send a thousand of them for the cost of a single manned flight to Mars. That utterly negates the success rate issue. And for flexibility - the Mars probes are still going after more than two years. You never, ever get more than you paid for with a manned space flight.

  [*]Human space colonization is the future of humanity; learning to live in space and elsewhere

  Dubious.

  [*]Hardware repair

  More expensive than it is worth.

 

[Astronuc]

Unmanned space activities are not cheap. To the contrary; they are very, very expensive. The Cassini mission, $3.27 billion. The Mars Exploration Rovers, $850 million. That represents funding for thousands of scientists for several years plus an untold number of graduate students. Cassini, Spirit, and Opportunity have yielded incredible results, but at an incredible cost. Will Congress fund future endeavors such as these without the added impetus of being a pathfinder for human activities? I doubt it.

I believe the cost of Cassini has much to do with the destination, instruments and spacecraft . Certainly there is a high overhead cost - the contractors make a lot of money - because lots of folks are involved in the initial design/development/construction (of the spacecraft and launch system) and launch/operation. There is a lot of testing because once the spacecraft is lauched, there is no hands-on access.

ISS is extremely expensive ($10's billions), and that is in the neighborhood. I've heard estimates approaching $100 billion for a manned mission to Mars. And manned missions to further destinations are not practical at present (if ever). We can't even maintain the infrastructure we have on earth!

AI would be necessary for more productive robotic systems, however another problem with instrumentation is the radiation. The smaller the transistor size, the more likely it the microprocessor/RAM will get zapped by a stray cosmic ray. The only way around this is to build a craft with massive (and voluminous) shielding, or an active magnetic field, which can introduce other complications.

 

[D H]

I've heard estimates approaching $100 billion for a manned mission to Mars.

A manned mission to Mars is not practical at present. Zubrin is nuts. We simply do not know enough about the dangers to humans on Mars to send people there. NASA wants intelligent risk takers as astronauts, not suicides. Even if we knew enough about Mars, we do not have the wherewithal to get there and back using present technology. The logistics alone are orders of magnitude larger than anything done by any spacefaring nation. Barring massive international cooperation, the US will not be able undertake a human mission to Mars within the next forty years. So let's leave Mars out of this discussion.

In fact, let's leave human space flight out of this discussion entirely. Imagine for a second that Congress has banned governmental funding of human space flight activities. A future mission of the scale of Cassini or the Mars Science Laboratory would have to compete on its own merits against other science. How many such mission will Congress fund over the next forty years without the motivation that humans might eventually go to this planets? I posit none.

AI would be necessary for more productive robotic systems

Using AI for robotic missions is problematic for several reasons.

• The developers have to experts in both the problem domains and in writing AI software. Such people are very hard to come by and very expensive. One unexpected consequence of the government's penchant for awarding contracts to the lowest bidder is that domain-specific software is necessarily simple.
• Each little chunk of flight software has to run in a very predictable amount of time. AI techniques typically do not do so.
• Each little chunk of flight software must be demonstrably correct. Critical pieces must be provably correct. Demonstrating correctness of AI software against a wide range of inputs, including erroneous inputs, remains an open issue.
• Flight software is atrociously expensive. Shuttle flight software: One line of code per day. (We're doing an order or two of magnitude better than this now, but it still remains atrociously expensive.) Flight software, even for robotic vehicles, is incredibly small. AI software tends to be incredibly bulky.

However another problem with instrumentation is the radiation. The smaller the transistor size, the more likely it the microprocessor/RAM will get zapped by a stray cosmic ray.

That is a generic problem with space-qualified processors, and the problem is growing as transistor sizes shrink. Flight-qualified hardware is ten years or so behind state-of-the-art. The problem is even worse than that because avionics design limits the design of almost every system on a spacecraft . The details of a spacecraft 's avionics system is one of the very first things set in concrete as a result. Given the large lag between initial design and flight, the computers on a spacecraft are often fifteen or more years out-of-date by the time a vehicle flies.

 

[Poop-Loops]

I think you completely missed his point. He doesn't want to pay to send someone into space just so an astronaut can enjoy the view, when we can send a robot to do it at half the cost.

Right. And when it turns out we NEED to go to space, for whatever reason (evacuation, manned repair trip, whatever), what are you going to do? "Oh snap, we should have developed this more, because the experience in sending people to space would have been worth it. Oh well, let's just send a robot to space instead."

The point is, humans WILL go to space sooner or later. You're betting on later, simply because at this point it is expensive to do. Never mind that the better we get at it, the cheaper it will become.

 

[russ_watters]

ISS is extremely expensive ($10's billions), and that is in the neighborhood.

I toured a full-size mockup of a big component of the then termed "Space Station Freedom" (which would be launched by the Shuttle-C) when I went to Space Camp in 8th grade. Not including the cost of the canceled predecessors, the total cost of the 20 year project, ending in 2010 is $130 Billion.
http://en.wikipedia.org/wiki/International_Space_Station#Costs

I've heard estimates approaching $100 billion for a manned mission to Mars.

That is also likely close to an order of magnitude low. I've never seen an estimate that low, but this article critcizes the most common estimate of $1 trillion: http://www.thespacereview.com/article/119/1

If the article is right (tough, since there is no actual program yet), it will "only" cost about half a trillion.

AI would be necessary for more productive robotic systems...

I'm not sure why we would need more productive robotic systems. Much is made of the on-the-spot decision-making capabilities of humans, but again, the low cost of robotic spacecraft utterly negates this issue. A human can do things faster, but a human also must do things faster due to mission time constraints. Even at a pace of 10m per day (with daily decision-making pauses), after more than 4 years into their 3 month missions, the two rovers have done detailed surveys of paths of several km (7 and 11). They have easily accomplished as much as a similarly equipped manned mission of, say, 3 months (on surface) duration at no more than 1/500th the cost.

 

[D H]

Not including the cost of the canceled predecessors, the total cost of the 20 year project, ending in 2010 is $130 Billion.
http://en.wikipedia.org/wiki/International_Space_Station#Costs
That is also likely close to an order of magnitude low. I've never seen an estimate that low, but this article critcizes the most common estimate of $1 trillion: http://www.thespacereview.com/article/119/1

That article violates at least two of the three Wikipedia content policies: No original research (NOR) and Verifiability (V). The verifiable cost to NASA is $25.6 billion for the years 1994 to 2005. Adding future expenditures without accounting for the value of money is invalid. Adding 80% of the cost of the Shuttle program is highly invalid. The Shuttle program has a huge fixed cost. The interregna following the Challenger and Columbia disasters did not see a cost reduction in the Shuttle program. The author of the article did original research and did so badly to prove a point (which violates WP:NPOV).

Regarding manned trip to Mars. Yes, it would be well over $100 billion and could easily by ten times that. For that reason, there are no plans for a human mission to Mars and I am not advocating one. Please leave the issue of a human mission to Mars out of this thread. It is off-topic.

What I am asking you anti-human space flight people here to do is to envision a NASA without a human space flight program. Why is that so hard for you to do? Please stop comparing the cost of human space flight versus robotic space flight. Instead compare the cost and scientific return of a robotic mission to Mars versus a graduate student mission to the Andes. The robot loses, which is why the BNSC gets a paltry 0.035% of the UK budget. Robotic precursors are an important part of the overall exploration objective. Because of this, NASA's unmanned missions receive over 1/3 of the total NASA budget, or about 0.2% of the US federal budget. My conjecture is that that would fall to levels in line with BNSC funding shoulw the more vehement elements of the science community get their way.

 

[Astronuc]

I was part of a group of grad students (20+ years ago) looking at advanced propulsion concepts for missions to Mars and nuclear applications in space, including ISS. It seemed like every year, the space station got redesigned and decreased in size and scope. One thing about advanced propulsion (power) systems is that they have dual use.

The range of cost estimates for missions to Mars go back 20+ years. IMO, there is a lot of overhead built-in. I'd rather see a Manhattan Project approach, but the government seems to want to included everyone.


More recently, JIMO got scrapped along with NASA's Prometheus program, mostly due to cost and change in priorities.

 

[russ_watters]

That article violates at least two of the three Wikipedia content policies: No original research (NOR) and Verifiability (V). The verifiable cost to NASA is $25.6 billion for the years 1994 to 2005.

Do you have a source? Here's another article that says $100 billion (which includes $10 billion for the predecsssors) in 2000 dollars, which is pretty much in line with $130 B today.

Apparently, the $100 B comes from a 1998 GAO report (linked here): http://www.msnbc.msn.com/id/14505278/

In any case, I'm not sure why the cost to NASA would be more relevant than the total cost of the project. I'm not interested in what part of the space station costs, I'm interested in what the whole space station costs. The only way to compare different missions is to compare total costs.

Adding 80% of the cost of the Shuttle program is highly invalid. The Shuttle program has a huge fixed cost.

The cost of the program would get amortized over the number of missions, would it not? If the ISS weren't there, the Space Shuttle program would likely be canceled already.

Regarding manned trip to Mars. Yes, it would be well over $100 billion and could easily by ten times that. For that reason, there are no plans for a human mission to Mars and I am not advocating one. Please leave the issue of a human mission to Mars out of this thread. It is off-topic.

I didn't bring it up, but in order to properly compare manned to unmanned, we have to compare apples to apples.

...BNSC gets a paltry 0.035% of the UK budget. Robotic precursors are an important part of the overall exploration objective. Because of this, NASA's unmanned missions receive over 1/3 of the total NASA budget, or about 0.2% of the US federal budget. My conjecture is that that would fall to levels in line with BNSC funding shoulw the more vehement elements of the science community get their way.

This is a theoretical discussion, so while you may be right, that's a non sequitur and a matter for future politicians to work out.

My position is simply that robotic spacecraft are more cost effective. I'm not suggesting that we even should stop human spaceflight, much less that it would be politicially feasible to have the robotic without the manned.

 

[russ_watters]

One thing about advanced propulsion (power) systems is that they have dual use.

Well, the space-program-as-pure-research logic is ok with me as long as people realize what they are paying for.

 

[Laura1013]

This article made me think of this thread: http://www.space.com/entertainment/080328-robots-vs-humans.html. Cyborgs!

 

[turbo]

A well-timed article on the practicalities of sending astronauts beyond the protection of Earth's magnetic field.

http://www.space.com/missionlaunches/080331-radiation-shielding.html

 

[aquitaine]

We as a species have a choice: Stay stuck on this rock and probably go extinct eventually or expand into space and survive.

At the end of the cold war the US had a tremendous opportunity. The Soviet Union had collapsed, and so there wasn't a need for such a massive armaments industry. Instead of investing all these hundreds of billions of dollars in science and space exploration to better humanity, most of it remained locked up in the bottomless pit that is the military. The US spends $750 BILLION on it's military. $750,000,000,000, and that isn't even counting the hundreds of billions spent so far occupying Iraq (since that isn't included on the official budget). The military-industrial complex is not about defending America anymore, but rather it exists for its own self-perpetuation. If you don't want your tax payer money wasted, then push for an end to this military adventurism.

History will look upon this as a major missed opportunity; We had a chance and we blew it. With the end of cheap energy either already here or just around the corner, I'm not sure when we will have this opportunity again.