Self-Replicating Probe… REPRO

Robert Freitas designed a self-replicating probe, REPRO, designed to fly to another star-system and self-replicate. He based the design estimate on the “Daedalus” star-probe study by the BIS, but being a stellar orbital mission instead of a fly-by the mass of fuel was immensely more than “Daedalus”, a flyby probe. Roughly 10.1 million tons of fuel, and 600,000 tons of probe structure divided into 3 stages, launches and delivers REPRO to a nearby star.

The mass is dependent on the size of the payload – in this case REPRO’s business end masses 1000 tons. Freitas has become an expert on nanotech, especially nano-medicine, since then and believes smaller needle probes can be used instead massing a few grams. I’m not entirely convinced, but he might be right. Certainly nanoprobes have a lot of advantages, but they’re more fragile too, especially vulnerable to the cosmic radiation they’ll be bathed in en route. Perhaps the mid-range REPRO probes, massing a ton, might be the best approach.

Let’s stick with the 10,700,000 ton monster and see what that tells us. The fuel can be mined from a gas giant, and the solids from their moons. If we assume aliens are launching REPRO-style probes in order to cover the Galaxy in a network of monitoring stations then we can estimate just how much mass they’ve used in our solar system. Say a new REPRO arrives every decade – unlikely as we’d have noticed the fusion flames by now – and they take a decade to reproduce a new REPRO to launch off to other stars. Thus the solid mass can computed as:

M = 0.5*(T^2/(R1*R2))*Mp

T is the solar system’s age; R1, R2 the time between arrivals and reproduction, and Mp the finished probe mass.

Thus for 4.5 billion years of 600,000 ton REPROs arriving and breeding every decade there will be about 6 x 10^25 kgs consumed, or about 10 Earth masses. Sounds like a lot, but the Kuiper Belt was about 100 times heavier in the past – where did it all go? Perhaps we have an answer. About 171 Earth masses of fusion fuel would be needed from the gas giants, which would be a noticeable loss, changing the orbits of their moons. But it could also be star-lifted from the Sun without being missed.

Of course the 1 ton REPROs would’ve needed roughly 1000 times less and we’d be none the wiser. Or if we make the arrival rate somewhat lower to account for no telescope spotting a braking starprobe, then we get a similar result. One problem with there being 1 ton REPROs in such numbers (10^17) is why don’t we see their infra-red glow? If they mass 1 ton, roughly 50kg/m^3 density then they’d be about 3.36 metres across, with about 9 square metres of cross-sectional area, their collective area is 6,970 times Earth’s cross-sectional area. That should be pretty visible in infra-red I’d think.

But consider: there’s about 10^16 kg of zodiacal dust interior to Earth’s orbit composed of specks about 100 microns across, thus about 5 x 10^24 specks and about 391 Earth areas of reflective surface – and we barely notice it. If the “cloud” of alien REPROs is out past Pluto we’d never notice their collective IR reflection as there’s quite a lot of plain old dust out there. And if Freitas’s nanotech needle probes are possible then there would be no way of noticing even a million times the number.


After I wrote this I noticed that Freitas assumes a 1,000 year reproduction time, which means the masses quoted above drop into near invisibility, especially if the arrival time averages about 1,000 years too. A 1,000 ton REPRO would be about 34 metres across and a fully loaded daughter probe (10,000,000 tons) would be several hundred metres at least. How many would be under construction at the present time? Assuming the REPROs are ‘immortal’ we’d have about 4.5 million original parent probes in the system and roughly as many under-construction.

If they’re not immortal and last maybe 10,000 – 100,000 years then REPROs might actively cannibalize inactive REPROs as there would be many more dead than alive. An ecology of scavenging machines might have evolved in the Outer Solar System and we’d never know about it. They might also use different fusion cycles – proton-Boron or proton-Lithium are other options – and not need to mine the Jovians at all, finding all their fusion fuels in the multitude of comets in the Oort cloud. I don’t think anything can be said dogmatically at this stage of our ignorance about such possibilities…

…but I think They’re here and They know about us.