Up in ARMs for No Good Reason

While there has been some slightly more positive discussion about the NASA Asteroid Redirect Mission since my previous blog post and SpaceNews Op-Ed, there have still been a steady stream of criticisms and suggestions of alternatives to the ARM mission being discussed lately. I’ve always been a bit of a knee-jerk defender of the underdog, and I still don’t think ARM is being given a fair shake by many in the space policy community, so I want to take the opportunity to respond to some of these criticisms and alternatives. And by respond, I don’t mean just dismiss–some of the suggestions have real merit, and could lead to ways to improve the existing ARM mission.


  1. We should be looking for asteroids not trying to redirect them — This criticism is that before sending a mission to an asteroid, we should be making more of an effort to find the vast majority of NEOs that we still haven’t identified. I actually agree that a strong effort at identifying more of the NEO population would be money well spent. As I pointed out in a previous post based on a talk by Josh Hopkins about LM’s Plymouth Rock mission concept, more knowledge of the NEO population and orbits gives us more options and can only make missions like ARM and others better over time. I don’t think this should preclude doing ARM as well, but finding ways to invest more effectively in this area would be useful. Some suggestions for how we could do better here range from traditional approaches such as funding a NASA asteroid hunting mission or  doing a competition for industry provided asteroid finders, to providing matching funds or a free rides for more commercial missions such as what Sentinel and/or the prospecting spacecraft  Planetary Resources and Deep Space Industries are developing, to potentially offering a bounty for detection and verification of new asteroids. It doesn’t have to cost billions, but spending a bit more money in this area is likely to make ARM better, and increase our odds of finding dangerous asteroids with enough time to do something about them.
  2. OSIRIS-REx is already doing this — I’ve heard people criticize ARM because the OSIRIS-REx mission is already going to be returning samples from an asteroid before the ARM mission is flown, and thus we don’t need another mission. I find this argument as silly as Obama dismissing the Moon with his “Buzz has already been there, done that” argument. Just as the Moon is a world the size of Africa, and you can only learn so much from a half-dozen landing missions, how much are we really going to learn about the millions of asteroids in the NEO population from a few hundred grams of materials returned by a handful of sample return missions? How much do we really know about the consistency of material composition even within one single decent-sized asteroid? What are the odds that even two C-type asteroids are going to be identical enough that additional samples wouldn’t be worth it?
  3. We need a large number of samples, not a large sample — I partially agree with this argument–diversity of samples is important, but so is having enough quantity to actually be able to do useful ISRU experimentation. The multi-teaspoon sized samples provided by current missions might provide you with some idea of the chemical composition of at least that part of a given asteroid, but learning how to mine asteroids (and if we can do so in a way that makes economic sense) is going to take larger samples. One possible compromise that could give you the best of both worlds might be a multi-lander/grasper ARM concept. Instead of having one big 4m diameter boulder grasping system, ARM could potentially do 6-7 smaller (~2m) separable grasper/landers, attached to an ESPA-ring like structure, and the spacecraft could also possibly visit more than one asteroid during the mission. As commenters have pointed out previously, there are actually relatively low-delta-V multi-asteroid tours that can be done that go from Earth to several interesting locations along the way before returning to earth. That way you can get sizable samples and variety, maybe 1-2 carbonaceous chondrite boulders, throw in a nickel-iron sample or two, and then one or two samples from another asteroid types. While this may sound a lot more complicated, NASA has already demonstrated multi-asteroid rendezvous with missions like Dawn, and building 6-7 copies of a lander grasper system would actually mean that a lot of the complexity is offset by larger efficiencies of scale–with a Prospector-like Option B grasper mechanism, you’d be making dozens to hundreds of most individual piece parts instead of the one or two that you often see for more traditional science missions. If I get more time, I’d like to flesh this idea out in its own blog post, but I wanted to get it out there publicly in case I can’t find that time.
  4. You shouldn’t pick the boulder just on its pluckability — One concern was that the boulder would be picked entirely from a standpoint of ease of extraction. I agree with this concern wholeheartedly, but NASA has already indicated that they were are planning to include at least a basic sensor suite to help with picking an interesting boulder, not just an easy one.
  5. You should bring the boulder back to Earth Orbit instead of DRO, because that’s just make-work for SLS/Orion — This one also comes up a lot. The fact that ARM is bringing the boulder back to DRO instead of LEO is seen as somehow indicating that this is all just a make-work stunt. But the more you study the problem, the more DRO seems like a reasonable choice. Spiraling in to earth orbit from escape velocity takes >5km/s of delta-V with a low-thrust system, on top of all of the other . This would require either a refueling or two, or a much bigger spacecraft (about 2-3x the size), and would take a really long time. High Thrust-to-Weight SEP stages can take 6 months to 1 yr to spiral out from LEO to escape. But with a 40-80 tonne boulder attached, the T/W ratio for the return spiral would be 5-10x worse, which would mean 3-10 years spiraling through the van Allen Belts. If you use aerocapture/aerobraking instead, with such a large mass, you would need either a large aerobrake, a lot of time, or something like the magnetoshell aerocapture technology we’ve been supporting MSNW on. I’m obviously not opposed to that last option, but this would be a non-trivial additional system development. Plus, even if you could magically snap your fingers and get an asteroid into LEO, there would still be challenges. An asteroid in LEO would be easier to visit but would also be a debris hazard (especially as you try to mine it and accidentally knock dust or rock chunks off or it), would have to deal with a much worse micrometeorite/orbital debris environment than it would in DRO, would be unlikely to have enough T/W to dodge a detected conjunction with other dead space objects in LEO, and would require constant propellant for reboost. It’s not an entirely impossible, but it’s not as much of a slam-dunk as some seem to think.
  6. Grabbing a boulder has nothing to do with planetary defense — This is one of the more ridiculous statements I’ve heard repeated by otherwise very intelligent people. The reality is that unless you’re going to use nukes, the gravity tractor is probably one of your better bets for asteroid deflection. And because the mutual gravitational attraction is proportional to the masses of the two objects multiplied together, there’s a big benefit for being able to increase the mass of the spacecraft using local mass. What better way is there to rapidly increase the mass of your spacecraft via in situ materials than to grab one or more big boulders off the asteroid?
  7. We should do ARM just minus the whole going to an asteroid and bringing a sample back thing — That’s like saying we should go to Mars but without that whole going to Mars thing. I think people are laboring under a false belief that the boulder grasping mechanism is most of the cost of ARM–it probably isn’t. The spacecraft bus and human spaceflight follow-on mission are likely a much bigger chunk, and NASA has already indicated they’d like to do those even if ARM was canceled. Canceling the grasping mechanism is unlikely to save you much at all–maybe the equivalence of a CRS mission or two, or a few months of SLS or Orion development. Spending the vast majority of the cost of the mission but without actually achieving useful exploration or ISRU development would be a waste. Why do people think that play-acting at being astronauts out at DRO without an actual useful mission for them to be performing is somehow more grown-up than doing actual exploration and potential ISRU research?
  8. We should skip the asteroid and go to Phobos instead — This is one of the best alternatives (not surprising considering the source–I have a ton of respect for Wayne Hale), and while I think it’s not the best option, I wouldn’t be heartbroken if ARM was refocused in this way. One of the selling points of ARM was that it is relevant to future Phobos/Deimos missions–the ARM spacecraft can and should be designed so that it can be refueled and “re-clawed” and used for another destination. The marginal cost of a Xenon tank and another copy of the claw is going to be trivial compared to the overall mission development costs, and there are tons of good reasons for an ARM-like mission to go to one or both of those moons. We didn’t explicitly analyze the case of grabbing a boulder from Phobos/Deimos, but a NASA Langley team did, and found that you could get a 1-2m diameter boulder off of them using the existing Option B hardware–notice this is the same size as the multi-lander/grasper concept mentioned above. But by skipping out on the asteroid first, you would lose the ability to test gravity tractor techniques, which could be important, and asteroids are also interesting in their own right. So I’m torn. I’d rather do both.
  9. We shouldn’t do anything that isn’t directly on the quickest path to Mars — I probably won’t convince Zubrinites, but it turns out we have this whole Solar System that doesn’t just consist of Earth and Mars. If manned Mars exploration was something we could do quickly, within NASA’s existing budget, or if there were no other interesting or useful destinations along the way, it might be one thing. But even the committee members who are advocating for this have admitted we don’t have the money to do a manned Mars mission in the next 25 years without significant increases in NASA’s funding. While it has been poorly marketed, Flexible Path wasn’t just about “doing asteroids first” or doing them instead of the Moon or Mars. To me the underlying point was that even if Mars is the long-term goal, we should find ways to do interesting exploration along the way to Mars, even if some of those destinations involve slight detours along the way. When you’re talking about a destination over 25 years out, acting like a 3 month delay is somehow insufferable is flat out ridiculous.
  10. We should just fly an SEP module to Mars and back instead of ARM — While the concept of skipping the asteroid and going straight for a Phobos or Deimos boulder return option actually made some sense–I think the concept of building a big SEP just to fly out to Mars and back is plain ridiculous. We’ve already demonstrated the ability to use SEP systems to do multiple rendezvous with celestial bodies, as mentioned earlier. SEP technology is likely going to shift so much over the next 25 years that the only good reason to spend a lot of money building and flying a demo SEP system now is if we’re using it for something useful like ARM. Building an ARM-class SEP system and just flying it around with no greater purpose seems like a waste to me. And as mentioned previously, you’re not actually saving that much money by ditching the whole grasper thing.

I could go on, and there are other positive suggestions I could provide, like using a COTS model on the SEP module to make something that gets us the experience we want while still being commercially relevant. But I wanted to provide some more thoughts for the ongoing conversation. ARM may not have very good odds of being funded to completion, but it’s not because the arguments against it are actually all that sound.

The following two tabs change content below.
Jonathan Goff

Jonathan Goff

President/CEO at Altius Space Machines
Jonathan Goff is a space technologist, inventor, and serial space entrepreneur who created the Selenian Boondocks blog. Jon was a co-founder of Masten Space Systems, and is the founder and CEO of Altius Space Machines, a space robotics startup in Broomfield, CO. His family includes his wife, Tiffany, and five boys: Jarom (deceased), Jonathan, James, Peter, and Andrew. Jon has a BS in Manufacturing Engineering (1999) and an MS in Mechanical Engineering (2007) from Brigham Young University, and served an LDS proselytizing mission in Olongapo, Philippines from 2000-2002.
This entry was posted in NASA, NEOs. Bookmark the permalink.

11 Responses to Up in ARMs for No Good Reason

  1. Nothing wrong with capturing and moving meteoroids into cis-lunar space, IMO. The best place to put them would be at EML4 and EML5.

    But a meteoroid deposition program really has nothing to do with NASA’s $8 billion a year human spaceflight related program- unless it also includes a meteoroid pyrolysis program for the production of oxygen, water, and hydrogen plus a serious LOX/LH2 propellant depot program.

    Of course, it would be a lot quicker and cheaper to get water and oxygen and hydrogen from the lunar surface than from a NEO meteoroid.


  2. Heinrich Monroe says:

    We should be looking for more asteroids to make ARM better? No, I don’t think that’s the point of looking for more asteroids.

    So OSIRIS-REx is doing this, but we need to sample more asteroids? Well, gee, then do more OSIRIS-RExs. And how many asteroids is ARM going to reach? Oh, you mean that when ARMI is done, we can budget for ARM XI and XII? We can have a whole flock of big boulders in that DRO? That’s nonsense. The take-away from asteroid fondling isn’t enhanced by diversity. No one is going to want to pluck another boulder and fondle it.

    Picking an “interesting” boulder? Yeah, I’ll bet. The only thing interesting about the boulder they pick will be it’s pluckability. That’s the single metric for success. And good luck with that. BTW, a loose boulder laying on the surface is most likely not geologically associated with the main asteroid. So the boulder we pluck is likely to be even more of a random sample than we would have wanted to believe. The role of the rest of the asteroid was just collecting it.

    “Grabbing a boulder has nothing to do with planetary defense”. Yep, you’ve got it. Your argument that follows make little sense. ARM isn’t a gravity tractor experiment. Yes, what ARM would do is test out an SEP for manhandling large masses, but the need for such a test is skepticism about Newton’s Laws. We don’t need to push a 100 ton boulder to know if we have the thrust to push 100 tons. If it doesn’t have a boulder to push, the SEP will just push itself a lot faster. Actually, I think that sending such an SEP to Mars with a nominal, and well understood and controllable payload would be a fantastic proof-of-concept. If we were going to ship random boulders to Mars, ARM might be a good test. But we’re not.

    “Why do people think that play-acting at being astronauts out at DRO without an actual useful mission for them to be performing is somehow more grown-up than doing actual exploration and potential ISRU research?” Because touching a rock isn’t the actual useful mission. Everyone at NASA understands that. The science community understands that. At best ARM is Mars-forward by sending humans into deep space, and that task simply doesn’t need a rock. According to Gerst, there is no “play acting” in that. You mean that if you don’t touch a rock you’re “play acting”? How quaint. We should all have rocks to touch. I guess all those ISS crews are just “play acting”, since they don’t have rocks they can touch.

    “To me the underlying point was that even if Mars is the long-term goal, we should find ways to do interesting exploration along the way to Mars, even if some of those destinations involve slight detours along the way.” I think that’s very smart, but ARM isn’t really interesting exploration. ARM is an excuse for interesting exploration. It’s wishful thinking for interesting exploration.

    I won’t bother with your other bullet points, because it sounds like you’re straining to come up with justification for ARM anyway. Yes, ARM is kinda “cool” and it’s sorta “hard”, but beyond that, it’s only purpose is to artificially create a relatively implementable and low risk nearby use for an SLS and an Orion.

  3. George Turner says:

    Would it make sense to just fly a close pass of the asteroid shortly after an impactor (or small explosive) struck it, kicking a nice random material sample into the path of the collecting satellite, which then slowly flies through the debris with an open trash bag? Given the extremely low gravities involved, and that the ejected material would be sorted by proximity to the blast and not material composition or size, you might get a more representative sample and save weight over a grab or scoop method. The method should lend itself to multiple rendezvous encounters on a single mission.

  4. Andrew_W says:

    Your #3 doesn’t look much like a take down of a criticism, rather it looks like an alternative that makes a hell of a lot more sense than ARM.

    A multi asteroid landing and sampling mission, with the samples returned to Earth, that would provide a great scientific return.

  5. johnhare johnhare says:

    Geatano moment, “Waa waa, I thought of the enhanced gravity tractor first and they stole my idea, waa waa waa. Thought of it once does not equal first, or best, or detailed enough to be useful, but I do enjoy the fantasy. 🙂

    Seriously. Is there any chance that a rock will be selected that could use the tether concept for sending mass back to cislunar space? A fast tumbling rock with a beanstalk past astrosync could throw multiple samples home. We discussed this on this site a few years back. After the primary mission of sending samples, the spacecraft could use the stalk to get to the next target.

  6. Pingback: Allgemeines Live-Blog ab dem 20. April 2015 | Skyweek Zwei Punkt Null

  7. Bob Steinke says:

    I’ll make a comment on item #4. Even if NASA includes a sensor suite, pluckability will still wind up as the top criteria. I worked at JPL during the development of the MER mission (Spirit and Opportunity). The engineers had veto power over landing sites based on engineering risk of landing successfully. It will wind up the same way with pluckability. On the other hand, look at how much science we’ve gotten from the “boring” Meridiani plain. We have so little access that any access at all is a bonanza.

  8. Hop David says:

    #9 is likely the argument that will keep humans from going much further than LEO.

    For growth of space infrastructure, there has to be revenue streams. Right now there is no economic incentive for having humans in space. No revenue streams from Mars colonization.

    I can hear the Zubrinistas wail “A full blown Martian civilization will generate tons of revenue!” For decades they’ve been placing all their bets on step 3 in the South Park Underpants Fairies’ scheme. The step #2 Zubrinistas continue to ignore is establishing such a civilization relies on nonexistent revenue streams.

    Can Planetary Resources or Deep Space Industries enjoy ROI? It’s a long shot but at least within the realm of possibility. Giving them an accessible DRO test bed would give them better odds. As would developing a heavy duty SEP with the smarts to rendezvous and retrieve inanimate objects not designed for docking.

    IF PR or DSI do enjoy ROI, growth of space infrastructure will follow naturally. Then prospects for Mars colonization will go from “Not A Chance” to “Inevitable”.

    Asteroid retrieval is being dissed as a distraction that will postpone sending humans to Mars. I would say the Send-Humans-To-Mars-Now crowd are the distraction. They are postponing the settlement of space. Maybe they’ll postpone space settlement forever.

  9. Paul451 says:

    Lose the astronauts-to-DRO component and you slash the cost of the program. Slash the cost and you can design ARM as a series of missions around common hardware, which spreads the initial development cost (the main cost of any robotic mission) out over more missions, reducing the per-mission price-tag. It also lets you incrementally develop the hardware, reducing development risks. And it allows you to incrementally increase the risk over multiple missions, starting simple (single visit, single-sample, low delta-v, low risk pluckable), and building up capacity as the hardware is gradually refined and confidence in working around asteroids grows (NEO tours, multi-sample-per-visit, multi-visits, building up to Mars moons and even main belt asteroids…)

    Suddenly ARM becomes a reasonable mission. From a technology development standpoint, an operational standpoint, and a scientific one.

    And it solves most of the criticisms (at least the sensible criticisms.) It increases the number of samples, it increases the number of bodies visited, it slowly increases the risk-taking precisely as understanding of the asteroids increases (you can start with safest-to-pluck but work towards risky-but-worth-sampling), and it is not a make-work program for astronauts. It even samples the Mars moons, throwing a bone to the Mars obsessives.

    “5 [….] Spiraling in to earth orbit from escape velocity takes >5km/s of delta-V with a low-thrust system, on top of all of the other . This would require either a refueling or two, or a much bigger spacecraft (about 2-3x the size), and would take a really long time. […]”

    Comparing the cost/difficulty of only the unmanned part of the two alternatives is, IMO, disingenuous. ARM’s astronaut visits requires the $3b/yr development of SLS and Orion, and probably won’t fly crew for 7-10 years. The cost of developing any additional unmanned system needed to bring the samples from DRO/L5/ESL1/wherever, back to LEO would be vastly (vastly!) cheaper than $21-30 billion.

    Hell, it’s actually hard to design an unmanned mission that would as expensive as Orion/ARM DRO rendezvous. Worried about debris in LEO? It’s cheaper to send a custom Bigelow-skin shell to shroud the whole asteroid in LEO than to send astronauts (**) to DRO. Worried about the cost of fuel to counter atmospheric drag in LEO? The cost of launching an extra 10 ton fuel tank for the SEP on F9 every few years is about the same as the weekly cost of SLS/Orion development.

    (** Unless Commercial Crew-to-ISS develops into CC-BEO after 2017. Which it would if decisions were not poisoned by politics.)

    “An asteroid in LEO would be easier to visit but would also be a debris hazard”

    I feel that this argument is only made by NASA post-hoc to justify the very specific capacity limits of Orion. By amazing coincidence the best place to do the mission just happens to be the furthest reach of Orion. I suspect if Orion could reach ESL1, suddenly the debris risk in DRO would have been an issue too. And if Orion/SLS could reach the NEOs themselves (the program sold to Obama), then, miraculously, any return to cis-Earth would have been just too risky.

  10. Bob Steinke says:

    @Paul451 “Slash the cost and you can design ARM as a series of missions around common hardware”

    If it’s unmanned it will be given to JPL who will unfortunately never design a series of missions around common hardware. (We’ve just had two wildly successful mars rovers so of course the only possible next step is a completely new rover design.)

Leave a Reply

Your email address will not be published. Required fields are marked *