≡ Menu

Micro ‘Bots’ to the Stars?

Debra Fischer (Yale University) takes a brief look at the next thirty years as part of a Discover Magazine 30th anniversary section, an appearance notable more for what Fischer doesn’t say than what she does. Any hint of how her radial velocity studies of the Alpha Centauri system are proceeding? I wouldn’t have expected any, I’ll admit, and Fischer says nothing about it, but the betting here is that we’ll have an announcement within the next year either by Fischer or Michel Mayor’s team either giving us a planetary discovery or sharply constraining the alternatives.

What Fischer does speculate on beyond the notion that we’ll detect life in exoplanetary atmospheres is that interstellar probes will eventually fly. You may recall Robert Freitas’ notion of interstellar probes loaded with artificial intelligence and as tiny as sewing needles, scattered into the galaxy in their hordes to investigate potentially habitable worlds. Fischer, too, likes miniaturization, which does so much to mitigate the huge propulsion issues:

Outside the gravitational field of Earth, we could launch robotic spacecraft to other destinations in our solar system. Further ahead I’d like to see tiny spacebots – smaller than your cell phone—travel outside our solar system to the nearest star system, Alpha Centauri. By keeping the mass of those spacebots low, we could more easily accelerate them. We could launch an army of these tiny bots and have them do what your cell phone does: take pictures and phone home.

Yes, and just maybe we could use them to create the kind of communications station at the Alpha Centauri gravitational lensing distance that Claudio Maccone envisages, using it to communicate at extremely low power with a comparable robotic relay at our Sun’s gravitational focus. That would set up tremendous bandwidth opportunities for tiny transmitters and allow valuable scientific studies to flourish.

Meanwhile, we’ll all keep speculating on the big question for the immediate future — when will the first habitable extraterrestrial planet be discovered? Greg Laughlin (UC-Santa Cruz) and Samuel Arbesman (Harvard University) have a go at this with a new paper that attempts a sociometric analysis of the question. The researchers create a metric of habitability that can be applied to already discovered planets and use a boostrap analysis to extrapolate discoveries into the immediate future. The prediction that emerges from this is near-term: The first Earth-like planet will be discovered (with high probability) by mid-year 2011. The method will likely be planetary transit or radial velocity (Debra Fischer’s Alpha Centauri work again comes to mind).

Will Kepler find the first habitable planet? Don’t count on it. For one thing, the next Kepler results we get will be of planets that are probably too hot to sustain life:

While the initial results of Kepler were released on June 15, 2010, the Kepler team has delayed publication of 400 of the most promising extrasolar planetary candidates until February 2011. Within this large pool of withheld candidates, it is virtually certain that some have radii that are observationally indistinguishable from Earth’s radius. It is likely, however, that because of the limited time base line of the mission to date, the Kepler planet candidates to published in February 2011 may be too hot to support significant values for H [habitability].

Laughlin and Arbesman re-ran their analysis using only those planets discovered via the transit method, learning that the method cannot determine a likely date of discovery because we have relatively few planets found by transits, and all rank very low on the habitability scale. But the authors’ habitability metric curve deployed on a larger population of 370 well-characterized known exoplanets continues to point to as early as May of 2011 and very likely by the end of 2013 for that first habitable planet. The method, fully described in the paper, is fascinating. What’s more, with target dates this close, we’ll have an early read on how prescient its authors really are.

Interestingly enough, the authors note that the habitability factor for most of the 370 planets in their study is zero, but of course Gliese 581d is an exception, recently examined by other authors and found to be potentially habitable. Laughlin and Arbesman disagree with the assessment, pointing out that the planet’s mass should be close to ten Earth masses. The paper describes Gl 581d’s ‘…possibly water-dominated composition more akin to an ice giant planet such as Uranus or Neptune than to a terrestrial planet like the Earth.’

The paper is Arbesman and Laughlin, “A Scientometric Prediction of the Discovery of the First Potentially Habitable Planet with a Mass Similar to Earth,” accepted by PLoS ONE (preprint). Re Gliese 581d, the paper is Wordsworth et al., “Is Gliese 581d Habitable? Some Constraints from Radiative-Convective Climate Modeling” (preprint).


Comments on this entry are closed.

  • yeti101 September 15, 2010, 12:02

    after keplers disastrous TED presenation i’m surprised to see more abuse of the term “earth-like”. Even if someone discovers an earth sized planet in an earth like orbit around a star like sol they still wont know if its habitable until we get spectro analysis of the atmosphere. Which is more likely to be nearer 2025 than 2011.

    such a discovery should be termed a terrestrial planet in the HZ. Not “earth-like”.

  • Carl Keller September 15, 2010, 12:31

    Professor Fischer’s comment, ‘Further ahead I’d like to see tiny spacebots – smaller than your cell phone—travel outside our solar system to the nearest star system, Alpha Centauri,’ suggests that she is encouraged by her findings to date.

  • andy September 15, 2010, 13:12

    Regarding Gliese 581d, it has been pointed out that the radial velocity dataset for the star is rather too evenly-spaced, which leads to significant trouble in distinguishing the true period of the planet versus aliases. It is possible the correct period of Gliese 581d is in fact 1.01 days, which would throw all habitability considerations out of the window.

  • Scott G September 15, 2010, 13:20

    Regarding the miniaturization of space probes to achieve interstellar flight… I was taking a look at the simplified ideal rocket equation: mi = mf/exp(-vf/vex) — where: vf = probe cruise velocity [m/s], vex = rocket exhaust velocity [m/s], mf = probe cruise mass [kg], mi = initial launch mass [kg] — and came to the conclusion that a reduction in cruise velocity or increase in exhaust velocity gives a much more feasible initial launch mass (as opposed to reductions in the probe’s mass).

    For example, the launch mass required to accelerate a 1kg probe to 0.01c (~3,000,000 m/s) is 1.9424e+130 kg when assuming an exhaust velocity of 10 km/s but drops dramatically to 20 kg with a 1000 km/s exhaust velocity (achievable by an “advanced electrically-powered drive”, according to Wikipedia). On the other hand, if we chose to drop the probes mass to 0.001 kg (and kept the 10 km/s exhaust velocity) we’d only reduce our launch mass to 1.9424e+127 kg, still an untenably huge number.

    This makes me think that a beamed-propulsion approach is almost THE only way to go if we ever want to get a probe to another star within our lifetimes.

  • Mike September 15, 2010, 13:35

    Kepler detecting the first habitable exoplanet or the Kepler team publishing the findings before some one else’s discoveries? Of all the various exoplanet searches being conducted through the next 10 years Kepler will likely find the most number of habitable Earth-like worlds. But maybe not the first one.
    Perhaps one of the 3 teams examing Alpha Centauri will be. It would be a crowning achievement for Debra Fischer and her associates if they are the first ones who find a Earth-like world especially one right next door.
    The first extra-solar destination for those micro-probes maybe and a hell of an incentive to fund micro-probe development.

  • Mike September 15, 2010, 13:47

    Regarding my earlier post perhaps I should state potentially habitable Earth-size worlds in the HZ zones of their stars. Yeti101 does have a point about not labelling a planet habitable unless we know something inhabits it.
    So I will say potentially habitable exoplanets according to our understanding
    of what a habitable planet might be like going by our limited sample of one.

  • Parmanello September 15, 2010, 14:57

    I’m all for exploration, but filling the galaxy with what are essentially relativistic projectiles with a camera on them responsible? Filling the galaxy with trillions of 1kg probes travelling at 0.99c, each one with the equivalent kinetic energy of 132 megatonnes of TNT – if one of those things from an alien race hit the ISS it would be a disaster…

    I know space is vast and the chance of impact is likely negligible but I thought it was an interesting perspective.

  • Interstellar Bill September 15, 2010, 15:10

    I detect in all these discussions an ill-considered blitheness abouts the hazards of relativistic speeds, so allow me to expound on what to me is common sense. If I had time I’d submit a paper somewhere.

    Going much above 0.1c would require most of the payload mass to be low-density shielding deployed ahead of the vehicle. Without such aerogel bumpers a relativistic vehicle would probably be destroyed before leaving the solar system. It would communicate with Earth by a passive laser retroreflector. Any given speed will require a minimum amount of shielding mass per unit area cross-section, one that goes up with speed squared. I bet that a cell phone is too small, that 10kg of shielding will be required in order for 1 kg to survive light years of relativistic erosion.
    The erosion/collision hazard mandates needle-like vehicles with tungsten noses and the most important mass all the way in the back.
    Multiple vehicles would be launched in close succession, each staying in the momentum shadow of previous ones, reducing their hazard.
    Remember that 0.1c takes you to the Moon in 13 seconds and across Earth’s orbit in 3 hours, so you have to wonder what a science vehicle could discover on such a mad dash through another solar system. A one-kg payload could only have a small optical aperture.
    0.1c, or 30,000 kps, is a frightening speed, the kinetic equivalent of a nuclear detonation, with 10-micron dust specks becoming hand grenades.
    A 1,000 gee catapult would have to be 45 million km long and have a peak power of 300 billion Watts per kilogram of payload, and that’s at an unlikely 100% launcher efficiency. Picture a ten TeraWatt solar array, not very big in Mercury’s orbit (only 100km square). Whoever gets to aim that catapult would be master of the solar system. Same goes for launch lasers, which could easily fry targets on Earth. Sounds like a good topic for an SF novel.

  • Ron S September 15, 2010, 20:40

    I. Bill,

    The thing about a blog format is that it is easy to believe that nothing exists before the current or previous few posts. In fact these issues have appeared many times on this blog, and much interesting discussion ensued. I know that for myself, I tend to avoid repeating what I’ve said previously on a subject just because it again becomes current. Possibly others feel the same way. The regulars here are, I think, aware of the challenges of high velocities.

    I would go so far as to say that these will be nice problems to have since they require that we will have solved the problem of achieving high velocities with real spacecraft!

  • ljk September 15, 2010, 22:18

    Take this as you will: In 2008 Arbesman and Steve Strogatz analyzed how probable Joe DiMaggio’s 56-game hitting streak in 1941 was statistically speaking.

    Using Major League Baseball stats data from 1871 to 2005, they ran 10,000 simulations of the entire history of baseball on a computer. Their conclusions were that while long hitting streaks were not unusual (109 in one rare case), DiMaggio’s record in 1941 should have been one of the least likely times for it to have happened. But happen it did in our reality.

    Take that as you will when it comes to predicting the discovery of the first Earthlike exoplanet and here is the NYT article on the former story:


    As for microprobes strewn across the galaxy, how well would they hold up against cosmic radiation? How much and what type of shielding can one give to a small robot probe to keep it intact to the next star system?

  • Adam September 15, 2010, 23:03

    Hi All

    Space is really, really BIG. Collision hazards always seem insuperable when we talk about the kinetic energy of dust specks, but just how many are out there and how many is a space-vehicle likely to encounter? Space dust can be observed and measured in bulk because it gets in the way of starlight and we’re learning more about it’s actual nature via various probes encountering it as the stuff enters the solar system. Roughly 0.03 solar masses per cubic parsec forms the interstellar medium – 99% H/He and 1% other elements, and of the latter ~50%-10% has formed as dust specks typically ~100 – 10 nanometres across. One speck thus occupies a volume of about ~450-450,000 cubic metres of space.

    Each speck masses ~1E-18 to 1E-21 kilograms and encountering a probe doing 0.1c it’s kinetic energy is ~0.00045 J to 4.5E-7 J. Such specks concentrate that energy into a tiny area (1E-14 to 1E-16 square metres respectively) thus represent a huge intensity, but small total energy. For a circular cross-section 1 cm across (a needle probe) the encounter rate is roughly once every ~0.19 to 190 seconds at 0.1c (numbers depend on size, thus the range in times.)

    Those 1 milligram space sand-grains that we interstellar explorers dread, based on the ISM mass density, are – at the very worst – encountered roughly once every ~600 kilo-lightyears by the 1 cm wide space-probe. And that’s assuming ALL the dust is in the form of 1 milligram space sand-grains, which it isn’t.

    Of course a starship with 2,000 square metres of frontal area (eg. “Daedalus”) is a MUCH bigger target. It encounters a space sand-grain – at worst – once every 1,500 AU, which fortunately gives it plenty of distance to deploy counter-measures in.

    BTW a 10 micron dust speck packs ~450 J of kinetic energy at 0.1c relative velocity. Hardly a hand-grenade.

  • Michael Simmons September 15, 2010, 23:10

    I have a bunch of questions about the viability of sending interstellar probes.

    Ignoring cosmic rays, have we actually seen any object at all enter our solar system from outside?
    Asteroid, Comet etc travelling at any speed.

    I’m thinking that anything travelling fast will be eroded.
    How much matter in the form of small grains is blown away from stars due to radiation pressure?

    Anything with a complex structure (silicon or biological) might not last the distance due to its own natural radiation, cosmic radiation, Gamma-ray bursts, supernovas and material migration.

    If you froze seeds down to near absolute zero and sent them on a 10000 – 100000 year trip how many would be viable. I know the Svalbard Global Seed Vault thaws them out every so many year, plants and grows them, then re-freeze the new seeds.

    Could a Radioisotope thermoelectric generator function for 10’s or 100’s of thousands of years without maintenance?
    Could solar cells last that long even if they are inactive?

    I assume any probe would need to be completely dead for the trip since even a small power consumption would add up to a huge amount over a long duration.

  • James M Essig September 16, 2010, 12:21

    Hi Adam;

    Re “BTW a 10 micron dust speck packs ~450 J of kinetic energy at 0.1c relative velocity. Hardly a hand-grenade.”

    Good point!

    This is actually a wee little bit of energy. I unfortunely wiegh 425 pounds or about 200 kilograms and so when I merely hop from the bottom step from my porch to get to the my car during a thunderstorm, I probably impact with about 450 joules of energy. My shoes, my strained legs (probably should not be doing this at my wieght, Grins and Giggles!), and the grass covered lawn at the base of the porch, do not seem at all to mind this, and so a much more massive, extremely hard, and refractory shield, that is dense could stop 10 micron wide dust-specks.

    A water based shield, heavy liquid metal shield such as one made of liquid mercury or some other simmilar shield could diffuse the impact of 10 micron to 30 micron dust-speck . The liquid would simply flow back into the track left by the particle.

    High gamma factor space craft, if they are ever practical to build, need not necessarilly have a wide cross-section. On concept that interests me is the interstellar train where the front of the vehicle would have a shield that protects the train via a larger cross-sectional area foot print. The shield might have a conical configuration that is very sharp having a sewing needle like aspect ratio in order to provide for very shallow angled grazing incidence of impining dust specks.

    Some sort of neutron dense, stable strange, charmed, or botton matter might help with regards to the conical shield, but that is getting a little ahead of current and projected technologies for the next couple of centuries.

    Regardless, thanks for running the numbers on interstellar dust specks. Your analysis renews my strong confidence that we can solve the interstellar dust impact problem.

  • Eniac September 17, 2010, 1:07

    Note that as the probe gets smaller, the amount of ISM material encountered per cross sectional area stays the same. Shielding therefore becomes progressively more difficult as the probes get smaller.

    Note also that nuclear reactors are hard to miniaturize, for a variety of reasons. That goes for fusion as well as fission, I would think. For that reason alone, small self-propelled probes are probably not in our future.

    Ruling out self-propelled, what is next? Catapults are a (very long) stretch, as Bill has correctly observed. Sails, too, as they must be thin and are quickly eroded by the ISM.

    After all these thoughts, it seems to me that the old-fashioned massive Daedalus style craft may after all be our best bet for relativistic travel. I do think, as James points out, that a train configuration is best. In addition, the fuel has to be solid, non-volatile, and suitable for structural use if the enormous mass ratios necessary are to be achieved. Lithium deuteride appears to be the best of very few choices. That “train” will likely consist of thousands of “cars”, almost all of which are solid blocks of pure fuel. I would not want to be around when it detonates, which is a possibility with lithium deuteride. Luckily, nothing short of a nuclear bomb will serve to trigger such a nuclear detonation, or so we are told.

  • andy September 17, 2010, 20:26

    Ignoring cosmic rays, have we actually seen any object at all enter our solar system from outside?
    Asteroid, Comet etc travelling at any speed.

    There are detected interstellar meteoroids. A lot of them apparently come from the direction of the star Beta Pictoris.

  • andy September 18, 2010, 13:16

    Yeti101 does have a point about not labelling a planet habitable unless we know something inhabits it.

    No no, that’s the difference between “habitable” and “inhabited”. A habitable planet is surely one on which life could have evolved, regardless of whether it did.

  • Adam September 19, 2010, 5:52

    According to Wikipedia’s review of the ISM dust its size-number distribution n(r) is proportional to r^(-3.5), which means 1/3 the mass is in particles ten times bigger. As the ISM distribution peaks at 100 nanometres, then 1 milligram/millimetre sand grains have a ~(1/3)^4 times lower mass fraction than the peak size i.e. 1%. Translated into collision frequency it means a sand-grain is encountered ~4.64 times less often than the worse case I outlined in the previous post i.e. about ~0.1 lightyears apart for a “Daedalus” size probe. I think. If anyone wants to check, that’d be appreciated.