Hacker News new | past | comments | ask | show | jobs | submit login
NASA to use consumer Android smartphones in new satellites (nasa.gov)
168 points by anigbrowl on Aug 27, 2012 | hide | past | web | favorite | 100 comments

I worked on this project for a few months. We ran everything like a typical agile startup with weekly iterations and a lot of testing "MVPs". Really cool project. Also its all going to be opensourced! http://open.nasa.gov/plan/phonesat/

That's awesome! Is following @NASA_Phonesat the best way to find out when that happens?


What was your radiation testing like? My understanding is that there's a lot of skepticism that this thing will run for very long. NASA uses hardened processors to make sure radiation from outside the atmosphere doesn't affect electronics and using an off-the-shelf SoC built on a tight whatever nm processes has never been done before for fears of early mortality.

The only exception I can think of is laptops on the ISS and on the space shuttle, but those were within the shielding of the craft. On top of that, those laptops are hardened against radiation. I believe they took off the shelf Lenovo Thinkpads and added some shielding.

These are probably low earth orbit satellites, still protected by the geomagnetic field and their orbits wouldn't spend much time in the Van Allen belt.

To be safe, I imagine they gave the phone a tinfoil hat.

I'm guessing that the weather balloon got high enough where it might experience some radiation. I'm curious if they experienced any watchdog timer resets during that test flight.

This is exactly what I was thinking about after the Curiosity launch. Space exploration desperately needs to iterate faster, and tiny satellites running commodity hardware/OS designs. Android seems perfect for this, considering how easy it would be to deploy updates as .apk files, develop USB peripherals, etc..

The article doesn't mention anything about radiation hardening, though. Are they just not worried about it, because the idea is to make the satellite as cheap as possible?

By the sounds of things, these are temporary satellites not supposed to last for very long, so the long-term aspects of radiation damage might not be such an issue.

Why is it important to make satellites which are cheap to build? Isn't the total cost dominated by the cost of bringing it to orbit anyway?

No, the satellite is usually the expensive part.

Also, the reasons for not making launches cheaper has been "why is it important to make launches cheap? Isn't the total cost dominated by the cost of the satellite anyway?"

Also, ground processing and all can get very expensive because it requires so many people and clean rooms etc and can be delayed. It depends if you think it's part of the satellite or the launch cost.

Overall you need to start somewhere if you want cheaper spaceflight. If there are cheaper satellites, then that's a big incentive for developing cheaper means of launching (and launching much more often too!).

It's partly a chicken and egg style thing. Launches are extremely expensive and far between and the big standing armies and facilities sit unused but have to be trained, maintained and paid. It's hard to break that circle, but it will be done, probably through some totally new technology that is first seen as just toys.

It cost's over 20k to launch a Cubesat it often costs far less than that to build them. Build costs and R&D can be hard to separate, but the replacement costs for many generic satellites are often well below the launch costs.

Check out the pics on the side - they're using a balloon, I don't know if that gets it all the way there but hobbyists do it - one of the more popular hn'ers does it and has blogged about it a but but I forget who.

Doesn't look like anyone's reached space yet but:


Interesting discussion on some guys who reached 100,000 feet:


Edit: actually they're launching via rocket -

"Three NASA PhoneSats systems (two PhoneSat 1.0's and one PhoneSat 2.0) are scheduled to launch aboard the maiden flight of Orbital Sciences Corporation's Antares rocket from NASA's Wallops Flight Facility at Wallops Island, Va., later this year. "

Yes, and I am currently working on GAGA-2: http://blog.jgc.org/2012/01/gaga-2.html, http://blog.jgc.org/2012/01/gaga-2-distance-rtty-test.html, http://blog.jgc.org/2012/01/gaga-2-mounting-flight-computer.... The vague plan is that GAGA-2 will be all about lots of cameras pointing in all directions. GAGA-3 will be about precision landing probably using a guided parafoil.

None of this is about orbit. Orbit == high velocity and altitude. Although high-altitude balloons regularly go about 40km that's still in the stratosphere and nowhere near space and the velocity is nowhere near what's needed.

These phone-based cubesats are a good idea. There have been plenty of phone-based balloon flights and there's no reason not to use them in satellites if you have some way of recognizing failure because of cosmic rays.

Don't sell yourself and other high-altitude-ballooners short—40km is almost halfway to space and is an amazing achievement.

Remember, everyone, space isn't that far away—if they put in a highway, you could drive there in about an hour. It's easy to think about space being very far away from us and untouchable, but it's actually quite close and just out of our reach. Keep reaching, fellow humans.

> Don't sell yourself and other high-altitude-ballooners short—40km is almost halfway to space and is an amazing achievement.

He's just being accurate. 'Ballooon reaches spaaaace!!1' articles upset H-A-B-ers more than most.

The average Cubesat launch price now is under $40k, but it keeps going down. Nasa now has $20k launch opportunities: http://www.spaceref.com/news/viewsr.html?pid=38673 and some others have under $20k as well it seems: http://www.interorbital.com/

I imagine that part of the benefit to making small satellites is the ability to deploy a lot of them. Also, it's a good PR move when public perception is that NASA is bloated and wasteful. (Please, nobody give your opinion for or against that; I don't care :P)

Expensive launch encourages expensive satellites. Expensive satellites encourage expensive launch. It's a vicious cycle.

I think it's important to make everything cheaper, even if you do get the most bang for your buck from reducing the cost of launch.

The other advantage in having NASA lead the way in the cheap satellite area, it will be easier for others to take advantage of cheap launches when they are eventually available. And possibly help create demand.

Space is a unique and unusual place. Usually the stuff that goes up there is also unique and unusual (hyperspectral radiometers, synthetic aperture radars, rovers). So it's expensive to engineer.

The cubesats (this is one) were developed as a way to prototype ideas and to get a larger segment of people involved in space activity. They are squeezed in to other launches. They're a very special case.

The phenomenon of cheaper, experimental missions is fractal. There are technology demonstration missions at many different price points (up to the 100s of millions of dollar range) that may have fairly modest science goals, but they're trying to show that some exotic bit(s) of tech (autonomous landing, laser communications) can work in space.

They are tiny as well, so in one launch you could put a LOT of them up there! For some kinds of operations, it's better to aim for 95% reliability and put quite a few in orbit, than to go to the expense of 99% reliability.

When I think of cell phones in orbit, the thought which comes to mind is "space junk." Putting lots of small objects in orbit cheaply sounds like a good way to increase the odds of something big, expensive, and possibly manned suffering a catastrophic collision.

In addition to the other comments they also can launch multiple at the same time for redundancy.

It would also be about size. If they can fit 10 satellites in the space used by a typical one it would make launch a lot more cost-effective.

Sorry if I'm being too nitpicky here, but NASA isn't actually innovating on size, here. They're putting everything in the standard 10cm^3 cubesat envelope. And they can fit a whole lot more than 10 satellites in the mass+volume of a typical one. And they can basically stow away on a rocket whose primary mission is one of those larger satellites.

Edit: For innovation on size, check out KickSat

I wonder how they address radiation hardening and other space specific environmental drivers in this hardware.

Radiation, such as cosmic rays, would tend to cause transient bit errors in memory that would cause the phone to crash or otherwise stop working. The article mentions that they have a circuit that automatically re-boots the phone when it stops transmitting a signal, so it can recover from these problems.

Wow. That's surprisingly un-NASA. I wonder if they have qualified the MTBF? For this kind of solution to work, it'd have to be at least a couple hours.

Actually, a lot of smaller things that go up just go into reboot cycles when they detect that kind of error.

Because at the end of the day, "Radiation hardening" just means, "Less likely to experience one of these errors but oh hey a solar flare is inc..." Satellites just sort of float until their orbit decays, they are de-orbited, or they are lost.

You wouldn't do this for a spy satellite you need to send you images of the sky in realtime for end-of-the-world missile guidance, but for something that's doing science missions it's probably way more important to make something light and cheap that a russian mission can drop out the back of its launch.

Maybe in this case it's cheaper to just send them up and measure it directly? It won't survive much TID, either, but with such a short mission and such cheap hardware, hopefully none of that matters.

We have short operating lifetimes and use orbits that minimise this problem.

Is radiation hardening really that important when you're still this side of the Van Allen belt? My understanding is that it's not, but IANARS.

Nice link, thanks. Very detailed.

Even airliners start having issues at 30000 ft.

Use a shield.

Of lead? At $10,000/lb launch cost?

Lead wouldn't work anyway - it would make things worse actually.

Lead is good for gamma (and X) rays, but very very terrible for cosmic radiation which is protons and neutrons (nucleons).

For those you want water.

It works like this: For photons you want them absorbed, so heavy metals are great.

But it's not possible to absorb nucleons. Instead you want to steal their energy. If you use lead they simply bounce loosing hardly any energy.

Instead you want very light elements - hydrogen especially. Then when they bounce all the energy gets transfered to the hydrogen.

Lead can make things worse because the nucleon can actually fission the lead atom emitting secondary radiation that is even worse than the original.

If you are shielding from electrons you can get Bremsstrahlung radiation if you use lead see: http://en.wikipedia.org/wiki/Bremsstrahlung#Radiation_safety

In summary: Lead for photons, water (or other light materials like wax or plastic) for everything else.

You've written an incredibly interesting comment addressing something I had no idea about, thank you.

You can try it out a bit.

Shoot various size coins at each other and watch how they interact, then try a coin against a wall.

If you have a very slippery surface try making a line of 3 coins (or more if it's super slippery) to bounce against each other.

I bet they'll be stuck on 2.1 forever.

I know this is sort of a joke, but the Nexus S shipped with Android 2.3 Gingerbread, and it's been updated to Android 4.1 Jellybean as of July 20th.

Is the update for the Nexus S OTA? I haven't gotten mine yet. In fact, I haven't gotten the 4.0 (ICS) OTA yet. I heard they released a version but pulled it... I have yet to see anything beyond 2.3 for my Nexus S.

We got it OTA.

...This raises questions about that other article that claimed everything had to be specially hardened to withstand radiation, no?

I had always assumed that was the case and the article text was obvious, but if we can launch off the shelf Android devices into space with no problems then why are the other NASA projects utilizing such out of date hardware?

As soon as you cross the Van Allen belt you're exposed to the full blast of the solar wind, and standard electronics start failing. You can reach orbit without crossing this belt, so my understanding is that in low Earth orbit you're still shielded enough from the solar wind for it to not be a deal-breaker.

Assuming that you're referring to the stories about the Mars rover, Mars isn't geologically active enough to have its own equivalent of Earth's Van Allen belt, so the Martian surface is directly blasted by radiation originating from the solar wind. Any computer hardware going there will almost certainly need to be radiation hardened.

tl;dr: LEO != Mars

Even in LEO (say, ~200 miles), the South Atlantic Anomaly can cause significant SEUs to standard hardware. But as you say, once you get higher than this, it gets really bad.

I'm guessing that these spacecraft have none of the mission requirements that most others have. They're not doing a flyby mission where they have to work during the flyby, or doing anything actually critical (such as provide comms/science, or use propulsion). The system is designed to reboot in the case of problems, which

The closest to a mission goal I can find in the post is "This approach allows engineers to see what capabilities commercial technologies can provide, rather than trying to custom-design technology solutions to meet set requirements." So the only goal is to see _if_ they work.

Trust me, if other missions could get away with less shielding, they would. That's one reason why RBSP[1] is being launched, so we can better understand the radiation environment in Earth's orbit, so future missions aren't undershielded or sent up with unneccessary shielding. But I doubt that most big-budget missions could get away with no shielding.

[1] Shameless plug/full disclosure: I have been working on this project for the last 2+ years, and am now one of its flight controllers. For updates, you can check rbsp.jhuapl.edu or http://spaceflightnow.com/atlas/av032/status.html

Surely the phone is going to be radiation-shielded inside the satellite. I see no critical contraction here...

Right but if we can just simply "radiation-sheild" off the shelf devices, why didn't we do that instead of custom building a robot made of specifically radiation hardened parts that are also much more expensive?

Why was the most important consideration by far for the processor in the previous NASA project was designed from the ground up to be radiation-hardned, but yet we can launch this no problem?

I just don't get it. I also don't appreciate the downvotes. I'm not trolling or looking to start an argument. I don't have dog in this race, I'm genuinely confused.

The answer is that modern consumer memory packages have come a long way from core memory. Standard ECC tech is extremely resilient to bit-rot and can recover in cases where older memory technologies would have failed catastrophically.

And in the case of real memory corruption, i.e. where cosmic radiation overwhelms the ECC controllers ability to correct the bit rot, well .. "reboot and retry" has been a standard for years, and will continue to work as a strategy as long as it is considered by the software architects of the system.

I think it has a bit to do with how mission critical something is. With the Mars rover, if it fails that is a lot more money and time down the drain than if a small phone satellite does.

As I commented on a Curiosity article, I hope they eventually tend more towards using lots of commodity hardware running instructions in parallel and working off probability rather than any one processor having to be perfectly deterministic.

I'm a huge fan of the possibility of heavily shielding off the shelf parts. It seems logical to me to try to incorporate it into unmanned craft.

However, I'm guessing there are probably people who have simulated this and it probably hasn't worked out to be very economical for past missions. I think part of the problem is that the level of shielding required would really limit your payload to orbit. It's probably going to take a huge amount of material seeing as even the Space Shuttle couldn't act as a perfect buffer between computers and radiation.

From a NASA article[1]:

"Designers also found out that laptops would crash when the shuttle passes through the "South Atlantic Anomaly," which is an area where the magnetic field draws in to Earth, again offering less radiation filtering for spacecraft flying through it."

Completely speculating here, but I would imagine that part of this experiment is to get operational experience with using standard grade electronics in space.

From the OP's article, the satellite has:

"...a watchdog circuit that monitors the systems and reboots the phone if it stops sending radio signals."

Anything with a human or with a high price tag attached to outages can't work with that caveat. If DirectTV's satellites did that, people wouldn't stay signed up for long. Imagine a Mars rover having to reboot every half hour... not very convenient.

But perhaps NASA might want to know if they can cheaply build "disposable" orbital experiments... i.e. ones that can complete their missions in a few hours. The sort of reliability where you just have to deal with the odd 30 seconds of rebooting every so often could be alright, especially if the payload can be stowed on a cargo ship and simply shoved out an air lock like the SuitSat[2].

Again spitballing, maybe they will find that things work out more or less OK if they send up five moderately shielded Raspberry Pis and they "vote" a la the Space Shuttle's flight computers[3]. If so, maybe that opens up space experimentation to universities with smaller budgets to squeeze onto resupply/manned launches that happen to have a spare square foot of space.

[1] http://www.nasa.gov/mission_pages/shuttle/flyout/flyfeature_...

[2] http://science.nasa.gov/science-news/science-at-nasa/2006/26...

[3] http://history.nasa.gov/computers/Ch4-4.html (search for "vote" to get to the specific spot detailing how this works)

I'm not convinced. Shielding against cosmic rays would require very heavy, thick lead walls to get any significant protection.

See my reply here: http://news.ycombinator.com/item?id=4437792 Lead is terrible for cosmic rays, instead you want water.

Wonder if Apple will sue to get an injunction against the launch. :)

With all respect, this isn't the kind of comment I've come to expect from HN lot on a serious article like this. It belongs to Verge/TechCrunch/Engadget/Gizmodo, or the the articles about the court ruling (those from yesterday).

Fact: In the current tech-landscape it's hard to think about innovative Android-devices launched without anticipating Apple filing anti-competitive lawsuits.

You reap what you saw. It's kalled karma. In the tech-sector, Apple's main perception is now largely as the lawsuit-company.

Really? At least in the circles I run in (bay area startups) Apple is perceived as the company everyone blatantly copies.

I thought I was the only one. The whole sports team like us versus them of the Apple/Android "wars" is getting old. I grew up in the Ford vs Chevy part of the US, I don't see much difference between the two and I'm getting sick of the whole litigation "notnews" being paraded around constantly to get pageviews.

Its getting to the point I don't even look at Android or Apple posts here because I know I'll start seeing the same memes yet again.

I think its time to stop going to HN, I don't see this improving anytime soon. I know, internet drama and all that but for once I would like to be able to read about the tech without the snark.

Actually the comment had a purpose and was trying to raise awareness. The fact that it made you angry perhaps shows that it worked? Perhaps it might make you reconsider your opinions about your favorite company.

I don't think a comment on HN, of all places, is going to do any raising of awareness of Apple's patent position.

Reddit, Youtube, MSNBC, etc., maybe. HN? C'mon.

Also, to get the comment, you already had to have that awareness. It wouldn't make any sense to someone who doesn't know about the Apple vs. Samsung lawsuit.

I'm all for raising awareness for what you think is right (even though Apple is my favorite company, while I disagree with if not most then many of what they're doing and how they're doing them), but this comment was like the typical Christian (or for my case, Muslim) missionary: "Hi, how are you doing today! Nice weather, isntit? Oh, and by the way, love Jesus(Allah) a bit more why don't you? He's a great fellow, Jesus is." - most of the time hurting their cause more than they do to help it.

It wasn't the OP's intent, I'm sure. All of us get angry some time and mix things up and comment on unrelated articles. I've posted more unrelated/stupid comments that I care to count.

And by the way, I'm not remotely angry. I have a lot of arguments about the subjects and have argued with many about it (not passionately, but all of us were reasonable and not typical fanboys). But I don't want to see them reiterated over all HN stories, that's all.

Worse. It’s a typical YouTube comment. Utterly devoid of content and insight. Also incredibly dumb.

You're right, it was a drive by comment, I'm big enough to admit it. My irritation has been building over software patents for years and it is leaking into a lot of my posts, not always a good thing to have anger cloud your thinking.

I did learn one incredibly useful thing from this thread however, was that Boeing has a patent on specific orbital mechanics, you know, the kinds of stuff people use to write papers about in the scientific realm for centuries for free transmission of knowledge. I suppose the next step is actually patenting scientific discoveries, like publishing a theory of Quantum Gravity via the USPTO instead of via Nature or the physics preprint service.

Not unless the satellites are rounded of course :)

The rounded joke is the easy one to make, of course. However, Apple did lose against Samsung on that count. There must be some sort of herding mentality that people go through, where people in general grab onto sound bites, instead of taking the time to do a deep dive into the issues. It usually happens in politics. On HN, I would hope for more.

*herding. On HN, I would hope for better spelling.

They'll be bared from entering into a rounded orbit.

You jest but a Boeing patent on a specific satellite manuever had the effect of decommissioning a satellite.


Perhaps patents will be the last thing that saves human-driven cars.

You can get an injunction against self-driving cars that can do the newly patented three-point-turn (its not prior art because now it uses a COMPUTER system!), but you can't stop a driver from doing it!

Patents definitely foster progress by the looks of it.

They also have to ensure there aren't any objects that might bounce back during lift off. :)

And you thought Apple was famous for their "product launches"

Looks like the Mythbusters will be invited for this next product launch and they will bring a lot of explosives with them.

This is an interesting experiment. I think it creates an opportunity to discuss the whole cost picture. COTS alone will only achieve minor cost reductions relative to the whole enterprise.

Testing costs were orders of magnitude more than the cell phones. They ran tests with Thermal Vac, vibe and shock, and weather balloon. Each test required time at expensive test facilities and hundreds of hours.

Launch and operations will require additional hundreds of hours of labor to prepare and collect data. It will also require time on expensive range assets.

For example, a launch readiness review, where they determine if everything is ready to go, would involve a dozen engineers for a few hours. This meeting alone would cost more than the all the phones they bought to test on the ground and fly.

I'm sure it will take a long, gradual process, but I hope that using cheap components will lessen the cost of failure, so we can spend less on testing, too. Now if only we could reduce the cost of launch....

Yeah, cost control is a perennial problem. I think it is so hard to solve because, it seems to me, that any solution must address interlocking technical, cultural, commercial, and policy & political concerns all together.

Will all these cheap satellites (from NASA or hobbyists, there are a couple of them on Kickstarter) cleanly de-orbit themselves or are we just adding up to the space junk at this point?

They are all put into orbits that decay to prevent this problem.

Let's just hope Apple haven't patented smartphones in space already.

Wonder how they solve the orientation problem.

Not so much with Android, but so much of Nasa's technology has spread to the private sector it's nice to see them using this to their advantage.

From reading the article, there is no reason why Android was chosen specifically. So this experiment could have been done with any modern smartphone out there, they just happen to choose a couple of android phones. Is that right?

I've got nothing to do with this project and only know what I read in the linked article but the seemingly obvious reason they'd use Android phones, especially Nexus phones, is that they run on a nearly fully open source stack (minus some binary drivers), so you can go as deep into the internals as you want without resorting to jailbreaking and/or a lot of reverse engineering.

Wouldn't you still need to root the phone? The stack is definitely more open than on iOS though.

Yeah we do have to root the phones but given the opensource aspirations of the project itself, Android was the most suitable.

Other NASA pages about this project seem to emphasize the "Open source" qualities of the project. There's also language like "with the goal of allowing anyone with space ambitions to launch their own satellite".

For instance, most other major options (elephant in the room being Apple), have onerous terms for their accessory development kits which would preclude that. Last I checked, Apple even keep the Dock port pinout "confidential" from people not in their accessory dev program, which would be a show-stopper since presumably this uses the dock/USB port for control.

I imagine the comment about having a "versatile operating system" has something to do with it.

My guess is the same reason ArduSat uses phone hardware.

All the sensors one might want on a Satellite (motions sensors, cameras, magnetic field detection) are all built into phones now. Android (or more to the point Android's Linux kernel) is specifically compiled to be compatible with most of that hardware. So less work for them as opposed to using another version of Linux.

What else would they use? It seems pretty obvious to go with Android.

NASA flies a lot with VxWorks. They could have ported it to the nexus.

At which point the satellites can no longer be open source. Perhaps more importantly, Android already runs on the Nexus S; VxWorks doesn't. It defeats the purpose of using commodity hardware if you start over from scratch on the software running it.

I agree VxWorks would require additional porting effort. I also think that Android was the right choice for this mission.

I'm not sure I agree with the implied high priority placed on open source. The relative development costs of android, real-time linux, or even a commerical RTOS are noise in the flight software development budget. Reducing the cost is a more important driver to making space more accessible.

The big benefit, if these android cube sats fly successfully, would be to retire a lot of risks - thus opening the door for future programs to fly with android.

Between the cheapness and flexibility of Android it makes it a pretty obvious choice, no?

Maybe it would be a huge experiment in redundancy fault-tolerant computing.

weird why not just use openembedded ? i forgot what the other one was called.

Applications are open for YC Summer 2019

Guidelines | FAQ | Support | API | Security | Lists | Bookmarklet | Legal | Apply to YC | Contact