PDA

View Full Version : Are Von Neumann probes doable without nanotech?



Noclevername
2018-Oct-13, 08:53 PM
The term for a macroscopically mechanical robot probe that can build copies of itself (https://en.wikipedia.org/wiki/Self-replicating_spacecraft) from raw resources, is a "clanking replicator". But can such a machine exist? Is a robot spaceship factory manufacture a working robot spaceship factory, too much for a mobile machine or swarm of machines? It would have to handle all aspects of mining, processing, and fabricating materials, all under space conditions. (Not, of course, using current Earth methods, but processes specifically designed or adapted to a space environment.)

It is near the pinnacle of speculative technology, second only to fusion power. We know it's physically possible, but can we actually do it?

cjameshuff
2018-Oct-13, 11:45 PM
Assuming that by "nanotech" you specifically mean self-replicating nanoscopic machines, why not? Why would it be possible to build replicating nanomachines, with all the limitations and difficulties of working at that scale, but not do it with macroscopic machines?

Humans have constructed an industrial base of clanking machines capable of producing all the machinery of which it consists. What are humans doing that only nanomachine replicators can replace?

Noclevername
2018-Oct-14, 12:19 AM
Assuming that by "nanotech" you specifically mean self-replicating nanoscopic machines, why not? Why would it be possible to build replicating nanomachines, with all the limitations and difficulties of working at that scale, but not do it with macroscopic machines?

Humans have constructed an industrial base of clanking machines capable of producing all the machinery of which it consists. What are humans doing that only nanomachine replicators can replace?

So, how do you put all that industrial system into an automated spacecraft?

cjameshuff
2018-Oct-14, 02:18 AM
So, how do you put all that industrial system into an automated spacecraft?

Make the spacecraft big enough?
You don't have to fit an entire active industrial base in the ship, just the stuff needed to build one. It's still going to be big, but you didn't specify any size limits.

Noclevername
2018-Oct-14, 02:31 AM
Make the spacecraft big enough? (snip) … It's still going to be big, but you didn't specify any size limits.

Fair enough.



You don't have to fit an entire active industrial base in the ship, just the stuff needed to build one.

But to build one you'd need an industrial base... that's what confounds me.

Solfe
2018-Oct-14, 02:48 AM
I always pictured these things working from comets and asteroids. Might as well use the whole object as your base. Once all the ships are built, you either don't have a massive object to get off of or you use whatever leftover waste mass as some sort of hull/shield for your next trip. At the very least, you have mass for some sort of mass driver removing the need to "leave" the surface.

DaveC426913
2018-Oct-14, 05:47 AM
I always pictured these things working from comets and asteroids. Might as well use the whole object as your base. Once all the ships are built, you either don't have a massive object to get off of or you use whatever leftover waste mass as some sort of hull/shield for your next trip. At the very least, you have mass for some sort of mass driver removing the need to "leave" the surface.
Though, counter-intuitively, it doesn't change the power requirements to leave.

Just because the entire asteroid is in the form of equipment or shields or fuel or whatever, doesn't mean it doesn't mass the same as the original asteroid it was made from, and therefore have the same gravity

(Unless, in the process of building everything, you eject a significant amount of it as waste material on an escape trajectory.)


Hm. There must be some sort of diminishing returns point.

At one end, a small spaceship needs to achieve escape velocity from a massive asteroid.
At the other end, an asteroid-sized spaceship needs to move an entire asteroid of mass. But it does not have to achieve any minimum velocity. Since it is a free-floating object, it can simply move away from that location as slowly as it likes. It's escape velocity is zero.

No, that's not right.
The least energy expenditure will always be achieved with the smallest possible mass being moved. ...Right?

Noclevername
2018-Oct-14, 06:53 AM
Though, counter-intuitively, it doesn't change the power requirements to leave.

Just because the entire asteroid is in the form of equipment or shields or fuel or whatever, doesn't mean it doesn't mass the same as the original asteroid it was made from, and therefore have the same gravity

(Unless, in the process of building everything, you eject a significant amount of it as waste material on an escape trajectory.)


Hm. There must be some sort of diminishing returns point.

At one end, a small spaceship needs to achieve escape velocity from a massive asteroid.
At the other end, an asteroid-sized spaceship needs to move an entire asteroid of mass. But it does not have to achieve any minimum velocity. Since it is a free-floating object, it can simply move away from that location as slowly as it likes. It's escape velocity is zero.

No, that's not right.
The least energy expenditure will always be achieved with the smallest possible mass being moved. ...Right?

The other half of the Delta V equation is energy. If the asteroid is, say, near a star, the VN can use a light sail, solar electric, solar thermal, or convert asteroid volatiles to fuel. And of course, use solar energy to run its factories.

If not, it needs an alternate source.

cjameshuff
2018-Oct-14, 12:40 PM
But to build one you'd need an industrial base... that's what confounds me.

No, you don't. More complex machines can be built with simpler ones, and components/materials with the largest footprint of required production equipment can be stockpiled to last you until their production machinery can be set up.

Seeds don't contain miniature copies of a fully mature plant.

cjameshuff
2018-Oct-14, 12:41 PM
Though, counter-intuitively, it doesn't change the power requirements to leave.

Just because the entire asteroid is in the form of equipment or shields or fuel or whatever, doesn't mean it doesn't mass the same as the original asteroid it was made from, and therefore have the same gravity

(Unless, in the process of building everything, you eject a significant amount of it as waste material on an escape trajectory.)


Hm. There must be some sort of diminishing returns point.

At one end, a small spaceship needs to achieve escape velocity from a massive asteroid.
At the other end, an asteroid-sized spaceship needs to move an entire asteroid of mass. But it does not have to achieve any minimum velocity. Since it is a free-floating object, it can simply move away from that location as slowly as it likes. It's escape velocity is zero.

No, that's not right.
The least energy expenditure will always be achieved with the smallest possible mass being moved. ...Right?

Where are you moving the asteroid? It can accelerate as slowly as it likes, but total delta-v required to reach another orbit around a planet/star is decidedly non-zero.

Solfe
2018-Oct-14, 01:32 PM
Though, counter-intuitively, it doesn't change the power requirements to leave.

Just because the entire asteroid is in the form of equipment or shields or fuel or whatever, doesn't mean it doesn't mass the same as the original asteroid it was made from, and therefore have the same gravity

(Unless, in the process of building everything, you eject a significant amount of it as waste material on an escape trajectory.)


Hm. There must be some sort of diminishing returns point.

At one end, a small spaceship needs to achieve escape velocity from a massive asteroid.
At the other end, an asteroid-sized spaceship needs to move an entire asteroid of mass. But it does not have to achieve any minimum velocity. Since it is a free-floating object, it can simply move away from that location as slowly as it likes. It's escape velocity is zero.

No, that's not right.
The least energy expenditure will always be achieved with the smallest possible mass being moved. ...Right?

Coming down to a planetary body and getting off again requires a lot of work and wasted mass to get back up. An asteroid allows you to dig in and use the resources in-sit while still moving. You don't need big rockets and aero-shells to get back off the ground. Telescopes can be simpler because they don't have a sea of air on top of them. No air means no weather, so you don't need that protection. If you are launch probes to many of the possible destinations in a system and you are taking years to build things, that small motion might get you to where you can do it for lower cost than a planet. Like Voyager's grand tour, that was only an option at the right time and place.

Over long periods of time, the asteroid could be rebuilt as the probes need. A mobile platform for exploration. Probably not an effective ship, but effectively a closet for all of you travel gear.

Noclevername
2018-Oct-14, 01:42 PM
At one end, a small spaceship needs to achieve escape velocity from a massive asteroid.


If you have the energy to construct a new Von Neumann probe, you'd have more than enough to reach escape velocity from an asteroid.

cjameshuff
2018-Oct-14, 11:00 PM
Coming down to a planetary body and getting off again requires a lot of work and wasted mass to get back up. An asteroid allows you to dig in and use the resources in-sit while still moving. You don't need big rockets and aero-shells to get back off the ground. Telescopes can be simpler because they don't have a sea of air on top of them. No air means no weather, so you don't need that protection. If you are launch probes to many of the possible destinations in a system and you are taking years to build things, that small motion might get you to where you can do it for lower cost than a planet. Like Voyager's grand tour, that was only an option at the right time and place.

Over long periods of time, the asteroid could be rebuilt as the probes need. A mobile platform for exploration. Probably not an effective ship, but effectively a closet for all of you travel gear.

This reasoning ignores the largest object in the solar system, which happens to be the strongest influence by far on the spacecraft for the majority of its journey.

Propulsive requirements for landing on/launching from a planet or moon are a fraction of the total propulsive requirements for getting from it to another location. Earth is the worst case of any body in the solar system with a solid surface, and the delta-v required to reach orbit is still only roughly equal to that required to get from that orbit to anywhere else. You need either big rockets or lots of time and energy to allow you to get there on ion thrusters.

Grant Hatch
2018-Oct-15, 03:02 AM
Are we talking in system or interstellar? In an interstellar scenario the nano replicators are obviously the way to go. The bare minimum mass required to travel between stellar systems and still have the ability to begin creating an industrial base once there.... the stellar distances require minimum mass in order to increase velocity between systems and shorten travel times to anything meaningful in a human lifespan given present drive technology. I'm reminded of the laser driven light sails proposed as a means to visit Proxima B. Oh dang, we still have to slow down....

Noclevername
2018-Oct-15, 03:08 AM
Are we talking in system or interstellar? In an interstellar scenario the nano replicators are obviously the way to go. The bare minimum mass required to travel between stellar systems and still have the ability to begin creating an industrial base once there.... the stellar distances require minimum mass in order to increase velocity between systems and shorten travel times to anything meaningful in a human lifespan.

I suppose interstellar. The point of VN probes, as I understand it, is to explore to places we can't see from here. Places in our Solar system, we can visit ourselves or send much cheaper non-replicating probes.

Grant Hatch
2018-Oct-15, 03:16 AM
Even in system a replicating probe might be cost effective. Would be nice to send them off to various resource rich places in system to begin making an infrastructure for us.... low delta v places best of course like the asteroid belt and select low G larger bodies....

grapes
2018-Oct-15, 06:24 PM
The least energy expenditure will always be achieved with the smallest possible mass being moved. ...Right?
Well, the goal is self-replication and dispersion, right? Why not, in the simplest model, have two identical structures fling each other in opposite directions?

Van Rijn
2018-Oct-19, 07:03 AM
Hello Noclevername . . . and everyone. I'm back (for a bit at least).


The term for a macroscopically mechanical robot probe that can build copies of itself (https://en.wikipedia.org/wiki/Self-replicating_spacecraft) from raw resources, is a "clanking replicator". But can such a machine exist? Is a robot spaceship factory manufacture a working robot spaceship factory, too much for a mobile machine or swarm of machines? It would have to handle all aspects of mining, processing, and fabricating materials, all under space conditions. (Not, of course, using current Earth methods, but processes specifically designed or adapted to a space environment.)

It is near the pinnacle of speculative technology, second only to fusion power. We know it's physically possible, but can we actually do it?

I'm of the opinion that we could probably build large self replicating systems with essentially current technology if there was a great enough need and enough money thrown at the problem. However, I don't think we yet have a good case for it - we have lots of humans to take care of the tricky bits on the planet, and fifty more years of technology could present much better ways to approach the issue. It's different from AI, where there are a lot of unknowns - in this case, we know how to manufacture, automate, and gather resources, though we will find better ways to do all those things.

The biggest issues I see are resource gathering efficiency (present day mining is not nearly as efficient compared to how biological systems gather resources), manufacturing requiring uncommon elements or especially complicated processes and the lack of human intelligence to handle special situations. I'd think if you were to try to build one, it would make sense to limit special requirements, even if the product wasn't quite as good (for instance, perhaps avoiding rare earths in magnets).

One idea I really like is not trying for 100% closure. For instance, chips (CPUs, ROMs, etc.) could be low mass and low volume but may also be more difficult to manufacture than motors, wires, structural beams, etc. This can greatly simplify the process, and it also has the great advantage that the replicators would have a limit - they couldn't take over the solar system or galaxy if there were reasonable limits on parts that couldn't be replicated.

Van Rijn
2018-Oct-19, 07:50 AM
I suppose interstellar. The point of VN probes, as I understand it, is to explore to places we can't see from here. Places in our Solar system, we can visit ourselves or send much cheaper non-replicating probes.

There's also the related idea of self replicating manufacturing systems (or a least high closure manufacturing systems). I have the opinion that significant human colonization of the solar system would only happen with high closure automated manufacture to do most habitat and other infrastructure building, and anything like terraforming (if we ever tried that) would require gargantuan space based industrial capacity probably only possible with high closure or fully self replicating systems. It needs to get relatively inexpensive to live on or above other worlds before many people would go there.

I could see humans making general decisions about what should be produced and where things should be placed, what worlds should be mined and where, but automated systems handling most of the details. That gets away from any general AI requirements, though specialized AIs might be helpful.

On space probes:

One thing about space probes is that deep space, aside from a bit of radiation, is a fairly benign environment compared to being in an atmosphere or in water. An advanced spacecraft designed for longevity with limited moving parts, shielded, radiation tolerant circuits and redundancy could potentially remain intact for thousands or perhaps even millions of years even without replicating systems.

I've had the idea of slow probes (mostly for some science fiction story ideas, but I think it also could be interesting Fermi issue related speculation). Picture a stable civilization many millions of years old. They might be happy to build space probes that can take tens of thousands of years to reach another star, in hibernation for most of the flight until they approach close enough to receive solar power. That would also greatly reduce propulsion requirements - the stars' relative velocities would be the primary issue.

Combine this with von Neumann systems: Perhaps you have a replicating spacecraft the size of a mile diameter asteroid, and it takes thousands of years to reach another system, where it looks for a fairly well differentiated larger asteroid (lots of different elements in one place but not too much gravity). In our solar system, Vesta might be a good choice. The replicator might go on to build a couple more replicators, and a lot more simple probes not designed to replicate themselves. Probes would go to other solar system, replicators would continue replicating in system if there were sufficient resources and those could build more probes, probes could eventually send back data about which other solar systems would be good replicator targets (probably not all would be).

Probes could sit around for literal ages watching systems and we've already seen some science fiction stories with that idea.

But the general idea here is that you don't need to limit yourself to just VN probes. You can have differentiated systems with replicators, or even replicator seeds that could (for instance) build vast mining/manufacturing complexes on asteroids which would in turn could build more seeds, non replicating probes, or other things (for instance, habitation modules for a biological species).