Reducing the size of a starship makes eminent sense, and as we saw yesterday, Alan Mole has been suggesting in the pages of JBIS that we do just that. A 1 kilogram interstellar probe sounds like it could be nothing more than a flyby mission, and with scant resources for reporting back to Earth at that. But by Mole’s calculation, a tiny probe can take advantage of numerous advances in any number of relevant technologies to make itself viable upon arrival.
Just how far can nanotech and the biological sciences take us in creating such a probe? For what Mole proposes isn’t just an automated mission that uses nano-scale ‘assemblers’ to create a research outpost on some distant world. He’s talking instead about an actual human colony, one whose supporting environment is first guaranteed by nanobots and, in turn, the robots they build, and whose population is delivered through the hatching of human embryos or perhaps even more exotic methods, such as building humans from DNA formulae stored in memory.
Let’s look at some of the factors the author lists, and bear in mind that we are trying to sketch out the shape of technologies that will have advanced in ways we can’t predict by the time such a probe is ready to fly, even if we allow it the relatively short time-frame (by interstellar scales) of fifty or sixty years of development before launch. From the “One Kilogram Interstellar Colony Mission” paper, here are the key points:
1) Increases in memory density show no sign of slowing. Mole cites small media memory chips that will soon carry two terabytes, but I’d point to Charles Stross’ fascinating discussion of ‘memory diamond,’ which sets theoretical limits on memory density by manipulating carbon atoms. If we need to pack vast amounts of memory into tiny spaces, the future is increasingly bright.
2) Within fifty years, nanotechnology may be able to produce tiny machines — nanobots — capable of complex tasks including self reproduction. The key question then becomes, can such technologies build humans? Mole recognizes the size of the challenge:
“Whether nanomachines can build full humans is unknown. It is physically possible — nature does it when a single fertilized egg cell grows into a human or animal. The DNA of a bacterium has been produced from stored ones and zeros in a computer. Granted, this required a full laboratory of equipment, but in five decades nanobots may be able to do it.”
3) I would feel better about the nanotechnology cited above if we took that fifty year restriction out of the equation, but even without humans ‘built’ by nanotech, we still have the option of sending embryos. Here the relevant citation is a 1989 Japanese project to incubate a goat fetus in an artificial womb, where the fetus grew to birth size but did not survive. Using vast numbers of human embryos on a colony ship, to be raised by robots at destination (robots that have themselves been built by nanobots), allows humanity to spread without large ships and without the need for hibernation (the large ships may be less feasible than the hibernation).
4) That Mole’s proposal is audacious is underlined by the fact that artificial intelligence may be its least controversial feature. Not everyone agrees with Ray Kurzweil that within three decades we’ll be able to essentially duplicate a human mind and run it as a program on a computer. But watching the trends in memory and recent work in brain architecture, including the Blue Brain Project, makes the prospect of uploaded minds at least possible. In any event, we are talking about running some kind of artificial intelligence on tiny CPUs that can manage the activities of nanobots as they build androids that go on to create a human colony. We’re in Singularity territory now, and in the nature of things, that makes predictions tricky indeed.
All of this grows out of a foundation of thinking that combines biology and silicon in interesting ways. Back in June of 1999, then NASA administrator Daniel Goldin spoke before the American Astronomical Society. It had been two years since he announced (in the same year that the Pathfinder probe landed on Mars) that reaching another star would be a new goal for NASA. That was startling enough, but Goldin went on to speak about a combination of lightsail technologies, artificial intelligence advances and hybrid systems tapping advances in biology.
It was an exciting time, even if the interstellar vision was quickly submerged in NASA’s more immediate goals and the ever present challenge of funding work in low Earth orbit. But Goldin’s probe — he described it as a space vehicle about the size of a Coke can — was meant to build itself by scavenging an asteroid, using the abundant supplies of carbon, iron and other materials such an object could provide. Mole’s paper reminded me of Goldin’s quote from that time:
“This reconfigurable hybrid system can adapt form and function to deal with changes and unanticipated problems. Eventually it will leave its host carrier and travel at a good fraction of the speed of light out to the stars and other solar systems… Such a spacecraft sounds like an ambitious dream, but it could be possible if we effectively utilize hybridized technologies.”
With Goldin as with Mole, the intent was to craft a starship without the need to push thousands of tons of payload, using the ability of technology to build and extend itself with local materials. In any case, we’re getting better and better at working with small spacecraft. Consider the Viking landers, each of which massed about 1200 kilograms (the Viking orbiter was 2300 kg). Mars Pathfinder’s lander came in at 100 kilograms, while the Sojourner rover itself massed only 12 kg.
Freeman Dyson laid out a concept for a 1 kilogram probe back in 1985 that set the stage not only for increased miniaturization but the fusion of biology with digital tech. Tomorrow I’ll get into some of Dyson’s ideas as a way of framing what Alan Mole is discussing, and then we need to focus in on the propulsion question. Getting anything — even something as small as a 1 kilogram probe — to another star is an extraordinary undertaking. But finding ways to leave the propellant behind can make it more feasible.
The paper under discussion is Mole, “One Kilogram Interstellar Colony Mission,” Journal of the British Interplanetary Society Vol. 66, No. 12, 381-387.
I’m reminded that Clarke noted that we usually overestimate technological development near term , but underestimate it long term. Charlie Stross’ skepticism about rearing the first generation humans from DNA (with which I concur) may suffer from underestimation of development.
Storage of human level minds in diamond offers some highly dense storage for minds. Consider that a minimal human mind might be stored in 1E-5 gm of diamond (1E11 neurons, 1E4 connections/neuron, 100 bits/synapse).
However, if the probe can assemble robots/androids and implant them with minds, why go the added complexity of building biological humans? The robots would be far more suited to the new environment. Robert Sawyer’s Mindscan satisfyingly has the mind-transplanted-to-a-robot body humans eventually live on Mars. But this mind scanning needn’t happen. It may be that we just need the probe to have a high intelligence (human-like or alien machine) and have its robot offsping develop more like biologicals – developing and learning. With machines, the whole habitat/terraforming issue just goes away and becomes an optional extra, in contrast to human colonization where we much create a terrestrial analog, even if it is like a worldship.
A pet idea that I’ve posted to Centauri Dreams a few times is a micro (1 gram? far less?) interstellar probe accelerated by a large structure in solar orbit (L2 or beyond) akin to the particle accelerators like CERN. Maybe a stream of these micro probes to a target, where they might coordinate with each other for relay communications or other purposes. But I think sending assemblers and embryos or DNA would make a difficult project even more so!
I agree that we are at a threshold of some remarkable
feats of technology. And that yes, it is more likely the first interstellar probe
will be measured in kilograms in mass, and not metric tons.
The big question is at what speed will they travel, and will they be capable of braking maneuvers of some sort.
Let us assume that you assemble a score of these probes
and pick say, 5 targets, at 4 probes to each target, (redundancy is a must)
If the probes cost what a robust probe fleet to the outer solar system would we get something like 10-15 billion. There is still the question of the cost to accelerate them, and to what speed. Are we to be exited about a probe that takes 50 years to arrive and then 50 more years to get a report back. I know that some have spoke of that as goal on this WebSite. But while we might
know that makes logical sense, as it lowers costs, it is a very hard sell, to
the taxpayer, and even private institutions. I think we have try to push
objects to high speed.
As far as I can see accelerating a probe to high C, is likely to be the larger cost of the mission, because an off-earth launch system would have to be built. The problem is protecting a such a probe from impacts is a probe design question and that coupled with a redundancy to each target is a solvable problem.
One thing, I don’t think there is a way to slow a fast probe without compromising the probability of success of the mission. There is no
propulsion system, solar sail solution that will work at high C. So at best your probe will streak by an alien solar system, (above it’s orbital plane probably) and go through it in a couple of hours. If your redundant probes
are scheduled to arrive months later, You might be able to make slight adjustments to their trajectory (another reason to send redundancy)
It sounds that what we want to do is send a vehicle (payload, apart from the propulsion system) with the lowest mass to: survive the trip; contain the minimum amount of information and machinery payload to establish a “base camp” with a sufficient amount of local energy and material; that can at least construct and power a communication system; exchange data with home or another base; sufficient basic machinery to put into effect consequent instructions to reproduce, explore, leave or colonize.
We cannot predict when we will be able to do so. But that’s seems to be the minimum requirement.
To start with we need to create a market and env. for space exploration. If we can integrate space and the universe into the education system to inspire the next generation of minds then we have a start. Simply having young minds engaged in thinking about space would be a start. From there we can breed the next batch of scientists / engineers / strategic thinkers.
Secondly, the UK needs an overall goal to focus minds and business. If we could suggest a short term goal such as the development of the ‘first space based solar power unit’ or a mission to the moon to assess mining potential etc we can bring the main players together. With manufacturing increasing here in the midlands it even makes sense to consider and plan for a space tech cluster to provide the infrasfructure for our automotive / aerospace / renewables industries (jlr / rolls royce). There are so many easy ways to start a national push towards space exploration. Just having the UK space agency in talks with strategic local authorities would even make a difference.
A really cool idea folks would involve dropping probes such as the one kg human fabrication probe down supermassive rapidly rotating black holes. Perhaps these probes would enter other universes or pop out in another location within our universe.
There may be some ethical issues with sending viable human embryos, however, if safe passage of the probe and embryos is possible, then we might be able to start one day what will go down in history as one of the greatest stories ever told.
What ethical responsibility do we have to ensure that we don’t “colonize” a planet with native life? While sending a 1kg probe that self assembles at the destination is an intriguing concept, it seems highly irresponsible to send it blindly without first establishing if there is native life at the destination.
Another risk is that we unleash an interstellar “plague”. Imagine the probe succeeding and then replicating further probes. What happens when a couple centuries down the road one of the second, third, fourth, etc. generation probes “colonizes” our system?
I apologize for the negative scenarios. I’m an engineer, it’s in my nature to play these games.
If it’s relatively easier to send out very small probes over interstellar distances – and that seems to be the almost self-evident case – then it follows that there may be some “alien Coke cans” hanging out in this very solar system, perhaps waiting for a suitable trigger. Or did I just simply come up with a decent scifi plot? :)
I believe it will be possible to offer my full paper by email shortly. Questions are arising here that were already covered there, at least in part.
Regarding microprobes I wrote:
“7.1 Smaller Probes
To follow this trend to its logical conclusion one may drop down a thousand times to a 1g probe. A 1kg probe is about the size of a grapefruit and 1g is only the size of a grape. Yet it appears that a probe could be that small.
Today 2 TB of memory weighs a tenth of a gram but for forty years memory has increased 100 times per decade. In five decades it will go first to 200 TB and .1 g, then 200 TB and 0.001 g, then 0.00001 g and so on. Clearly, memory and CPU can be done for a tiny fraction of a gram. Likewise nanobots of one billion atoms would weigh 2 x 10-14 g each. A payload one-thousandth that of the one kg considered above needs a sail of a thousandth the area or about a thirtieth the diameter — 9 m instead of 270 m. Beam spread will not be a problem if acceleration distance is a thirtieth as much and that can be done with thirty times the acceleration — 30,000 g’s. The existing beam generator could produce a thousand times the acceleration for a thousandth the mass, so thirty times will be no problem.
It may be possible to go further. Suppose that one atom can carry one byte of memory (perhaps one bit per electron, spin up or down.) 200 TB equal 200 x 10^15 bytes or atoms, say carbon atoms of atomic weight 12 AMU. Then memory masses
200 x 1015 x 12 AMU x 1.66 x 10-24 g/AMU = 4 x 10-6 g ”
[Note: I had not known of Charles Stross’ discussion of ‘memory diamond’, which seems to be a more advanced idea.]
“or four thousandths of a mg. With a CPU of equal mass, the total is .008 mg or just under .01 mg. If there are a thousand nanobots of a billion atoms each, that is another 1012 atoms, negligible compared to the 1017 used for memory. Thus memory, CPU and nanobots cost .01 mg. If another .01 mg is needed for batteries and solar cells and another for the structure and sail, then about .03 mg is needed for a colony probe at the absolute minimum. Thus one gram is possible and .03 mg is about the limit.
7.2 Costs
If a one kg probe costs $17 billion for infrastructure and $2.4 million per shot, then 1g would cost $17 million and $2400 per shot and .03 mg would be seven cents per shot. For $7 billion humanity could send dust mote colonies to all 100 billion stars in the galaxy. This is not to suggest this should actually be done but it is interesting that it would be possible. In contrast to a single generation ship that would bankrupt the solar system, miniaturization allows the affordable colonization of the galaxy.
7.3 Higher Speeds
With probes of 1g or less, higher accelerations would be possible and could yield speeds of .3 or even .9c. These would open a larger region of space using the same flight times. ”
Regarding braking: [In Andrew’s original plan]” After cruise, it is possible to slow the probe by reintroducing a current into the sail and using “friction” from interstellar magnetic fields. ” [In my modification] “As in Andrews’ plan, braking is via magnetic drag. Per Zubrin and Andrews, maneuver within the extrasolar system is possible using the magnetic sail and local magnetic fields.[15]”
Please email me for a copy of the full paper.
Thanks,
Alan Mole
RAMole@aol.com
I am very skeptical that we will achieve AI in the coming decades, let alone carry out successful “down loads” of the human mind onto any sort of silicon/carbon platforms. For me, the mind is too fleeting a thing to capture and in capturing it, we would ultimately end its essence.
This building of humans from stored information is another “tech” (I hesitate to call it tech as its somehow distasteful – for me) that seem somewhat unlikely. It’s uses are very questionable, ethically suspect and in the end in terms of humanity, pointless. For me, building humans at some far distant location out of basic building blocks is almost the worst form of imperialism we can practice and would just come to pass because we can say we “put” a human somewhere and say, “look we’ve made it there”. In actuality “we” would not have “made” it there at all. The spirit of the journey is the critical thing for me. Further, our journey into the beyond should be that in actuality, i.e. a journey consisting of people experiencing it.
I guess, ultimately what I’m saying it that once the human experience ends, as we “evolve” into that something other, I’m nonplused about what comes next. The human aspect will be gone.
@Thomas M. Hermann
Yes, there are certainly ethical aspects to consider. I have thought about this subject for quite a while.
As a supporter of the panspermia hypothesis (weird… yes, i know), i can’t help but also to factor in the possibility that there are no planets with environmental conditions that permit life which are uninhabited to begin with, which makes the subject even more critical.
On the other hand that paradigm also suggests there is already a natural exchange of viable microorganisms going on throughout the galaxy (and perhaps beyond). And if you want to put it on the more speculative side… this may actually be the rampart “interstellar plague” which maybe is already be in full progress for billions of years. And to be frank i think it HAS colonized our planet, which is why we are here to begin with. Image it is entirely in the realm of possibility we wouldn’t be discussing this matter at all in that case.
Its also entirely possible that the introductions of… foreign… organisms into a planetary biosphere has a beneficial effect. I am specifically thinking about the oxygen catastrophe on Earth some 2.4 billion years ago, as plants had poisoned the atmosphere which lead to a mass extinction until… well animal life took hold on Earth, metabolizing “poisonous” oxygen. I am not necessarily implying animal life didn’t evolve on Earth and was introduced later, but its a good example how one kingdom of life supports the other.
It means the introduction of foreign organisms is not necessarily hazardous, it may also SAVE a planets biosphere from extinction.
That being said both is true: there are certainly hazardous effects but, very importantly there are also symbiotic effects possible. We can’t tell for sure without an in-situ study.
I am recommending being less pessimistic about our own biosphere and its destructive capabilities. We are no “plague”, hence we wouldn’t think about ethical implications. Our genetic repository might actually be very, very valuable, containing solutions to many problems evolution has overcome on Earth (or… if i am right, also quite possibly on lots of other places).
That being said, molecular assemblers in any form, be it nanotechnology or genetically engineered biology or a fusion of both (a grey zone in any case) is no children’s toy. It is very, very powerful capability for both: either destruction or creation and as such has to be wielded very carefully, IF it can be wielded at all.
Actually, Kurzweil gives thirty years for an astronomically hyper-human mind…half that for a merely human one.
If we want to send microbes or Nano-construction entities into space and beyond as ‘seed pods’ perhaps we could encase them in colossal nanotubes with the ends sealed with water soluble materials such as salts. We could perhaps then weave them into sail or again in a larger blocks of salt to protect them and attach them to the sail. The sail could then be designed to fragment (weaker bonds in areas to make smaller sails) in the target system to perhaps to hit a planet or suitable body in a shotgun approach. The salt dissolves in the water or a suitable medium allowing the entities out.
I was re-reading Woodward’s book, “Making Starships and Stargates”, and it occurred to me that the same economic and energetic constraints as discussed here would probably apply at least somewhat even to a “warp-drive” or wormhole-traversing ship.
There’s no reason to think such a ship would be any less costly or energy efficient (quite the opposite, in fact). The trip would be shorter, but at the same time if you have the tech to send nano-factories and uploads, why not go miniature, anyway? It’s got to be easier and less energy-demanding to maintain a 1-cm radius warp-bubble than an 1 km one, after all.
Also, and this isn’t at all directed at this site, which I greatly enjoy just as it is, I haven’t found much worthwhile (i.e. knowledgeable, but also willing to engage with what he writes rather than refusing to even read the book) reaction or discussion about Woodward’s book, which I think is a shame–or perhaps reflects my not knowing where to look.
Gerry, yes, it has been some time since I’ve covered Woodward’s work, although you might be interested in this 2011 piece he wrote for Centauri Dreams:
https://centauri-dreams.org/?p=18076
Civilizationware could be downloaded upon arrival. Recieve costs magnitudes less then transmit. Advances in AI would be required only after arrival time.
Alan Mole:
Have you considered the power density of a beam that could achieve such acceleration, and how to protect the probe from being vaporized by it?
@ Eniac :
A graphene monolayer can handle ~10^7 W/m^2, so when we dope this for 100% reflectivity we get the ideal lightsail. However, we need to keep its temperature down to ~2000 deg K, which reduces the max permissible intensity to ~ 9*10^5 W/m^2. Then it would take about 21 days to sublime off one microgram. However, even with such an ideal sail, increasing its size eventually suffers from diminishing returns on the achievable acceleration, because the all-up mass increases – and way beyond 1 Kg, please note. A 5 Km diameter sail will have a mass of ~15 Kg, and will accelerate at about 750 gee with an incident flux of 18 TW.
@Eniac April 5, 2014 at 18:11
Alan Mole:
‘Beam spread will not be a problem if acceleration distance is a thirtieth as much and that can be done with thirty times the acceleration — 30,000 g’s.’
Particle beam divergence decreases as the velocity increases, in effect near the speed of light particles will tend towards self collimation as a consequence of its relativistic velocity. Think in terms of the twin paradox.
‘Have you considered the power density of a beam that could achieve such acceleration, and how to protect the probe from being vaporized by it?’
A parabellium bullet experiences ~200 000 g, however a sizeable loop may be a completely different kettle of fish, it does not scale well, those are huge forces taking place on a very flexible ‘ribbon’.