Beta

Slashdot: News for Nerds

×

Welcome to the Slashdot Beta site -- learn more here. Use the link in the footer or click here to return to the Classic version of Slashdot.

Thank you!

Before you choose to head back to the Classic look of the site, we'd appreciate it if you share your thoughts on the Beta; your feedback is what drives our ongoing development.

Beta is different and we value you taking the time to try it out. Please take a look at the changes we've made in Beta and  learn more about it. Thanks for reading, and for making the site better!

Android In Space: STRaND-1 Satellite To Activate Nexus One

timothy posted about a year ago | from the roaming-charges-will-get-you dept.

Android 103

An anonymous reader writes "In as little as a few days, the British-made Surrey Training, Research, and Nanosatellite Demonstrator (STRaND-1) satellite will begin transitioning its key systems over to a completely stock Android Nexus One smartphone that's been bolted to the bottom of it. The mission is designed to test the endurance of off-the-shelf consumer hardware, and to validate Android as a viable platform for controlling low-cost spacecraft. STRaND-1 managed to beat NASA's own 'PhoneSat' mission to the punch, which will see a Nexus One and Nexus S launched into space aboard the April test flight of the Orbital Sciences Antares commercial launch vehicle, the prime competitor to SpaceX's Falcon 9."

cancel ×

103 comments

Are the key systems of your satellite running? (0)

Anonymous Coward | about a year ago | (#43104483)

Well... I guess you better go catch them!!

I doubt that there's much cell reception in orbit, but it'd be cute if it could receive crank phone calls.

some one hack it so we can get free hbo (0)

Anonymous Coward | about a year ago | (#43105397)

Hbo

Re:Are the key systems of your satellite running? (2)

SnarfQuest (469614) | about a year ago | (#43108089)

Must be broken. All I get from it is a repeating message:

"Can you hear me now?"

Boiled alive (0)

Anonymous Coward | about a year ago | (#43104537)

If the radiation doesn't kill it the battery life will!

Re:Boiled alive (0)

Anonymous Coward | about a year ago | (#43105717)

Don't worry: I'm sure it has a solar charger.

Que random jokes (4, Funny)

Bobfrankly1 (1043848) | about a year ago | (#43104565)

I'd hate to pay the roaming charges on this.

It's stuck with a 2 year contract

At least it doesn't have to interface with iTunes

WHO FORGOT TO ADD TETHERING TO THE PLAN!?!

etc, etc, etc

Re:Que random jokes (4, Funny)

swanzilla (1458281) | about a year ago | (#43104727)

Don't forget the N900 crowd...someone was probably controlling low-cost spacecraft in 2009.

Re:Que random jokes (0)

Anonymous Coward | about a year ago | (#43105173)

Cue the fools who don't know the difference between "cue" and "queue" What manner of hybrid beast is this? Kway? Kweh?

Re:Que random jokes (1)

Anonymous Coward | about a year ago | (#43105517)

Cue the fools who don't know the difference between "cue" and "queue"

What manner of hybrid beast is this? Kway? Kweh?

Que?

Re:Que random jokes (1)

Aranykai (1053846) | about a year ago | (#43105753)

What?

Are you telling me the manager is faulty? Well, whats wrong with him!?

Re:Que random jokes (2)

wonkey_monkey (2592601) | about a year ago | (#43107599)

Ees no rat. Ees hamster!

Re:Que random jokes (0)

Anonymous Coward | about a year ago | (#43110887)

Is filligreeee Siberian hamster!

Re:Que random jokes (1)

denvergeek (1184943) | about a year ago | (#43106351)

Donde?

Re:Que random jokes (1)

Bobfrankly1 (1043848) | about a year ago | (#43110477)

Cue the fools who don't know the difference between "cue" and "queue" What manner of hybrid beast is this? Kway? Kweh?

Sorry, I must have been thinking about swimming...

Re:Que random jokes (1)

CannonballHead (842625) | about a year ago | (#43105597)

Opera was the first to do this in 2005.

Wow (4, Insightful)

P-niiice (1703362) | about a year ago | (#43104587)

This is so ridiculously cool. I'd never have imagined that cellphones would even be considered for such a thing.

Re:Wow (1)

Anonymous Coward | about a year ago | (#43105019)

Clearly, you have not seen Iron Sky.

Re:Wow (0)

Anonymous Coward | about a year ago | (#43105243)

Clearly, you have not seen Iron Sky.

Has anyone? I have it... I don't know what's stopping me from watching it... perhaps camp, nazi's and sci fi don't go together too well, not sure.

Re:Wow (1)

h4rr4r (612664) | about a year ago | (#43105313)

I watched it. I liked it quite a bit. Easier better than some movies I saw in theaters last year. Not the best movie of the year either.

Re:Wow (0)

Anonymous Coward | about a year ago | (#43105575)

Black to the moon!
Yes She Can!

Re:Wow (0)

Anonymous Coward | about a year ago | (#43105663)

It's pretty ridiculous, but I liked it. It's a good movie to watch with a group of friends while drinking/smoking. Or if you're just in the mood for something ridiculous.

Re:Wow (1)

Jhon (241832) | about a year ago | (#43106143)

I've rarely had a cell phone last more than 2 years with modest abuse. With the cost of getting equipment IN to space, the forces exerted getting equipment in space and the combination of hot/cold OF space, is it WORTH the savings if you need to replace the equipment fairly often? Also, what about 'space junk'. Wouldn't a better idea be fewer resilient longer lasting satellites?

That said, I agree, it is kind of cool, but I think that's just my knee-jerk geek reaction without much thought to how PRACTICAL it might be.

Re:Wow (1)

jacknifetoaswan (2618987) | about a year ago | (#43107073)

But you're missing the point...Android is a very extensible system that's not only low cost, but full featured, and you can build all sorts of new capabilities into it. Also, the guts of cell phones are extremely small and powerful, as well as cheap, and can be placed in a hardened container to protect them from environmental conditions (vacuum, thermal, radiation). Further, since they're so small, you can pack dozens of them onto a spacecraft, in multiple, independent, and redundant packages, such that if one, two, or three fail, you could have ten more to back them up. My Nexus 4 weighs 139 grams with battery. Packing ten of them into a package, with interconnects, might account for 10 kilos, including a radiation shield. My guess is that you'd be able to implement all associated infrastructure of many mission packages for far less mass budget than you would with a traditional spaceflight certified computer package, like the RAD750, which costs $200,000, and runs at 200 MHz.

Re:Wow (1)

Jhon (241832) | about a year ago | (#43108269)

"But you're missing the point..."

No... I got the point. Perhaps I failed in communicating mine.

I understand what you are saying. What I'm NOT getting is a cost-benefit break down. My question would be how long would the RAD750 last and how much does each (android or other) cost to get in to orbit. MOST of the cost is getting the bugger in to space. Once there, the LONGER it lasts, the cheaper it is per day/month/whatever cycle you wish to measure.

Re:Wow (1)

jacknifetoaswan (2618987) | about a year ago | (#43114567)

Well, the RAD750 is a hardened VME chassis, and, having worked with them in the past, a fully loaded VME can weigh 50 - 75 pounds, and that's not including any sort of I/O channel boards or shielding. If you can get the guts of an Android phone into orbit, or more likely, a cluster of them, for less than ten pounds, including shielding, you've saved yourself quite a bit on the mass budget. Considering that many spacecraft have multiple redundant computers, that mass savings could translate to the ability to launch a couple microsatellites, or to be able to carry additional instrumentation or propellant.

Re:Wow (1)

jacknifetoaswan (2618987) | about a year ago | (#43114577)

And by saying that a RAD750 is a hardened VME chassis, I meant to say that it runs in a hardened VME chassis.

Re:Wow (1)

Anonymous Coward | about a year ago | (#43108353)

But you're missing the point...Android is a very extensible system that's not only low cost, but full featured, and you can build all sorts of new capabilities into it.

So is bare Linux. The extra stuff Android adds to Linux isn't all that useful for running a satellite.

Also, the guts of cell phones are extremely small and powerful, as well as cheap,

So are many other things.

and can be placed in a hardened container to protect them from environmental conditions (vacuum, thermal, radiation).

You forgot vibration and high-G stress. Rocket launches are intense. Consumer gear isn't really designed for it.

Further, since they're so small, you can pack dozens of them onto a spacecraft, in multiple, independent, and redundant packages, such that if one, two, or three fail, you could have ten more to back them up. My Nexus 4 weighs 139 grams with battery. Packing ten of them into a package, with interconnects, might account for 10 kilos, including a radiation shield.

You're just guessing. You have no idea how much robust packaging sufficient to correct for all the deficiencies of a cellphone for this application might actually cost or weigh.

My guess is that you'd be able to implement all associated infrastructure of many mission packages for far less mass budget than you would with a traditional spaceflight certified computer package, like the RAD750, which costs $200,000, and runs at 200 MHz.

Here is a hint: $200K is peanuts. I know it's Wikipedia, and the cost data is a bit sparse, but this page can give you a rough idea:

http://en.wikipedia.org/wiki/Comparison_of_orbital_launch_systems

The cheapest cost-to-LEO listed is SpaceX Falcon Heavy (note: vaporware rocket, has not actually been launched yet) at $2,204 per kg. More mature designs with several successful launches are $3,784 to over $10,000, with most being over $10K.

So, even granting your previous guesstimate, if you need to pack in a bunch of redundant 10 kilo packaged cellphones? With rockets that exist, it'll cost you ~$40K to ~$100K per packaged phone. That'll eat your supposed cost advantage for avoiding a real radiation hardened computer right quick.

The only reason this satellite has a Nexus One is that it's a student "microsatellite" project. Build something extremely cheap, try out wacky and wild ideas, experiment and have fun. It doesn't have to survive long and failure is an option. (But note that it actually has a more normal flight computer that is the primary controller, with the Nexus as an experimental alternate.)

Actual cellphone HW is extraordinarily unlikely to revolutionize commercial or scientific research satellites. There's too much money on the line to entrust them to consumer-grade hardware which isn't even designed for that application.

What this student research has a small chance of demonstrating is that a relatively modern ARM SoC can be used in space without rad-hardening. The RAD750 isn't all that dissimilar to normal hardware, after all -- the CPU of that box is literally a PowerPC G3 (aka PowerPC 750), what Apple used to ship in G3 PowerBooks, PowerMacs, iMacs, and iBooks. The difference is that the RAD750 CPU chip has been revised with extra circuit features (more internal ECC protection, etc.) to make it much more resistant to SEUs (radiation events flipping bits), and is manufactured in a special rad-hard process (which is many nodes behind cutting edge). Both the circuit customization and special rad-hard process development cost a lot of money, so it's hard to do this kind of adaptation frequently for a market which is so small in unit volume, which is why they're stuck with an over-10-year-old CPU core.

If a project like this proves that it's possible to fly a cellphone and achieve reasonable reliability, what will actually be done is to design custom boards using off the shelf chips. Boxing up commercial phone designs running a commercial phone OS is, frankly, stupid. (While still being entirely appropriate for a wacky student project.)

(Or they may just go to Xilinx and ask them to space-qualify their Zynq chips, which feature dual Cortex-A9 plus FPGA fabric. Xilinx already ships a lot of space-qualified FPGA silicon; a lot of space hardware needs capabilities similar to ASIC but the budget isn't there to spin a real ASIC, so they use FPGAs instead.)

Re:Wow (1)

viperidaenz (2515578) | about a year ago | (#43111599)

Putting a Google Nexus in control of a satellite in space is pretty cool marketing though. Unless it fails in some catastrophic way and the whole thing falls back to Earth and kills some poor orphan in a 3rd world country on his way to his first day at school after being sponsored by a single mother-of-three.

Re:Wow (Not really.) (1)

Darth Snowshoe (1434515) | about a year ago | (#43106961)

Cellphones have not yet been engineered for space. They're not built to survive the radiation. They won't work up there for very long.

Re:Wow (Not really.) (1)

viperidaenz (2515578) | about a year ago | (#43111611)

Unless you put them in a box that blocks the radiation.

Sort of pointless (2, Interesting)

93 Escort Wagon (326346) | about a year ago | (#43104629)

Even in a "low-cost spacecraft" the cost of a consumer OS would be a trivial part of the budget - the difference between Android costing nothing versus the cost of stock Windows, iOS/OS X or Blackberry isn't particularly meaningful.

Off-the-shelf hardware, though - that would be a bigger deal. It's doomed to failure, but if somehow it could work that would be huge.

Re:Sort of pointless (2)

Doug Otto (2821601) | about a year ago | (#43104693)

Yeah, but at least they can move Android to another satellite without having to call MS.

Re:Sort of pointless (1)

K. S. Kyosuke (729550) | about a year ago | (#43104747)

Off-the-shelf hardware, though - that would be a bigger deal.

One might argue that a reasonably robust microcontroller that you can buy in your electronic parts shop would be cheaper and more reliable anyway.

has some advantages (2)

Chirs (87576) | about a year ago | (#43104777)

Running Android gets you a full-fledged OS that is also designed for low power consumption--but it's also open-source allowing for customization.

Re:has some advantages (0)

Anonymous Coward | about a year ago | (#43105395)

A full-fledged OS for human consumption isn't really that great of an advantage when you're basically making it a controller from an embedded system. In fact, it's a pretty big disadvantage.

Re:has some advantages (1)

jacknifetoaswan (2618987) | about a year ago | (#43107101)

Regardless, the kernel is massively extensible, and throwing out the UI, you've got a great system for which to do development, plus, it's free. Many of the newer real-time OS systems are based on Linux, but they're very heavy. Some RTOS development could make Android into a great embedded, real-time OS.

Re:has some advantages (1)

node 3 (115640) | about a year ago | (#43108209)

Then why not just use Linux? Or better yet, a proper realtime OS?

Re:has some advantages (0)

Anonymous Coward | about a year ago | (#43110815)

This.

For the cost of launching in consideration and all other factors... just use QNX or some such. Nothing like a kernel panic in linux to make your cheap satelite an out of this world paper weight.

Re:has some advantages (1)

viperidaenz (2515578) | about a year ago | (#43111621)

There is no weight in orbit.

Re:has some advantages (1)

jacknifetoaswan (2618987) | about a year ago | (#43114559)

Image size. I used to work with a real time Linux kernel called Red Hawk, made by Concurrent Computer Corporation, which was a layer that ran over Red Hat. Try as we might, we couldn't get the install image own below 500 MB, and that was without a GUI. Some of the smallest Android distributions are less than 100 MB, and they include a fully featured GUI. Also, Android is optimized to run in low power situations, on slower processors, rather than the quad core, multiprocessor systems we were running Red Hawk on.

Re:has some advantages (0)

Anonymous Coward | about a year ago | (#43109793)

Regardless, the kernel is massively extensible, and throwing out the UI, you've got a great system for which to do development, plus, it's free. Many of the newer real-time OS systems are based on Linux, but they're very heavy. Some RTOS development could make Android into a great embedded, real-time OS.

Why would you start from a heavy, non-RTOS customization of Linux optimized for human UI (aka Android) when trying to develop a lightweight Linux RTOS? Wouldn't it make more sense to start from an existing RTOS customization of Linux? (I hope you're aware that realtime is pretty tricky stuff to get right and it takes a lot more work than just tossing out stuff you don't need.)

Also, how are you determining that existing RTOS versions of Linux are "heavy"?

Re:has some advantages (1)

BitZtream (692029) | about a year ago | (#43107603)

Yes, but its power saving is still absolutely horrible compared to anything you'd actually use in space. Android is a 'desktop' OS for phones. Its only mildly concerned with power saving, in reality power on Earth is 0 cost free energy compared to power in space. It in no way compares to what software running on real sats does to conserve power. The whole scheduler is horribly horribly inefficient for those purposes.

Re:has some advantages (0)

Anonymous Coward | about a year ago | (#43111375)

Yes, but its power saving is still absolutely horrible compared to anything you'd actually use in space.

You know this how?

Android is a 'desktop' OS for phones. Its only mildly concerned with power saving, in reality power on Earth is 0 cost free energy compared to power in space.

Oh, I see, you "know" that by assuming things which are false.

Android is a mobile OS, not a desktop OS. It is highly concerned with power saving, because mobile platforms have extremely limited thermal dissipation and battery capacities. Power is so limited in most Android platforms that it's not uncommon for the LCD screen's backlight to use more power (on average) than the CPU.

Most Earth-orbiting satellites are solar powered, with battery storage to tide them over while in the Earth's shadow. Electric power is, in fact, effectively "free" for such satellites -- it's generated locally, and there's tons of it by cellphone standards (probably 2 or more orders of magnitude more, in most satellites).

Best of all, every orbit you get a guaranteed period of exposure to high quality sunlight (more W/m^2 than you'd get on the Earth's surface, and no weather). For this reason, satellite design engineers don't need to be concerned with saving every last joule of electrical energy at all times. It's not that power isn't a concern at all -- a larger power budget implies larger panels and batteries, which imply more weight and volume, and spacecraft designers want to minimize weight and volume -- but rather that once you know what the budget is (usually based on what systems you want to run simultaneously for a long period), you don't have to worry much about using less, because the panels and batteries can sustain it forever.

Also, it's not uncommon for satellites to include electric heaters, used to prevent temperatures from dropping too much while the satellite's in shadow. In fact, if you click through TFA's links and read a bit, you'll find that the students planned to deal with the Nexus' battery potentially getting too cold by deliberately running power-burning code on the phone's CPU, rendering all power saving features in Android irrelevant. In other words, waste heat generated by the CPU can even be a good thing in satellites.

Mobile devices, on the other hand, face far different constraints. If you make a mobile SoC continuously draw as much power as it's capable of, it'll drain any cellphone battery flat in an hour or less. Because users want a phone that lasts a long time on one charge, cell phone system design is all about going to insane lengths to optimize off-peak behavior. Idle power is minimized, and time spent in idle is maximized. Peak performance is used in incredibly short bursts between long periods of idle.

It in no way compares to what software running on real sats does to conserve power. The whole scheduler is horribly horribly inefficient for those purposes.

Your posting is horribly horribly inappropriate for anyone trying to learn about any of this stuff. There are legitimate reasons why using Android for space systems is kinda pointless, but you have masterfully managed to mention none of them.

Also, even in power constrained systems like a cellphone, the process scheduler is not a big deal. The real gains are found in userspace. The kernel does need to support a few features, but mostly it's about the application software. A scheduler can only save power by downclocking and/or shutting down idle CPU cores, and the scheduler can only idle cores when there's no code ready to run on them.

Apps which sit in a tight loop polling for events, or draw UI at a higher refresh rate than the display can physically support, or any of a gajillion other bad behaviors that waste CPU cycles? There's very little the kernel can do about them, short of refusing to give them as many timeslices as they can consume. That's not a great idea in most cases (it's very difficult to discriminate between a badly coded app and one which has a legitimate need to use as many cycles as it can get), so smartphone operating systems like iOS and Android tend to limit that kind of punishment to background processes which aren't presenting UI.

Space systems do tend to have kernels whose schedulers have a different design priority, but it's not about saving power. Instead, it's about ensuring that a handful of critical background processes (e.g. telemetry monitoring, orbital position correction) are guaranteed to receive timeslices on a fixed schedule, with a maximum permitted process wakeup latency after a sensor interrupt arrives. In other words, they're often soft- or hard-realtime kernels.

Re:Sort of pointless (5, Informative)

AikonMGB (1013995) | about a year ago | (#43104785)

Off-the-shelf hardware, though - that would be a bigger deal. It's doomed to failure, but if somehow it could work that would be huge.

Why do you say that? We use COTS hardware pretty much everywhere in our missions. It turns out that the radiation environment isn't really that terrible if you are below the Van Allen belts. Why pay through the nose, both in terms of dollars and in terms of horrendous lead times, for space-qualified parts when commercial, industrial, and automotive parts work just fine?

Re:Sort of pointless (0, Flamebait)

sl4shd0rk (755837) | about a year ago | (#43105117)

Why pay through the nose, both in terms of dollars and in terms of horrendous lead times, for space-qualified parts when commercial, industrial, and automotive parts work just fine?

Heh.. Proof right there that you're not affiliated with a US government agency. And, to answer your question, the reason is to keep the tax dollars funneling into the privatized black-holes* It's rampant.
[*]
http://rt.com/usa/blackwater-security-iri-report-300/ [rt.com]

http://www.aflcio.org/Blog/Corporate-Greed/ALEC-s-Funnel-Turns-Public-Dollars-to-Corporate-Profits [aflcio.org]

http://www.alternet.org/one-states-poor-excuse-funneling-taxpayer-cash-private-schools [alternet.org]

http://www.dailykos.com/story/2013/03/01/1181678/-Walker-s-Plan-for-Wisconsin-Public-Schools-Make-Them-Private [dailykos.com]

Re:Sort of pointless (3, Insightful)

AikonMGB (1013995) | about a year ago | (#43105241)

Heh.. Proof right there that you're not affiliated with a US government agency.

That's right, I'm not, and neither are the developers of STRaND-1. I'm not even American. The world is a big place, not all of it is funded by the US government.

Re:Sort of pointless (0)

Anonymous Coward | about a year ago | (#43105949)

What does the US government have to do with a European satellite? Either your reading comprehension is really REALLY bad, or you're in such hatred of the government that you use any excuse to bash it.

Tea party much?

Re:Sort of pointless (1)

denvergeek (1184943) | about a year ago | (#43106379)

Does the tea have mushrooms in it? Or are you talking about something else?

Re:Sort of pointless (1)

wiredlogic (135348) | about a year ago | (#43106875)

the radiation environment isn't really that terrible

Provided you are prepared to play dice with your mission longevity then sure, COTS is fine. Currently all successful COTS hardware in space has been incorporated into a system design tailored to handle the issues caused by operation in that environment. Efforts are made to detect latchup conditions and flipped bits that aren't done on ground hardware. An off the shelf smartphone has none of that implemented and will not work reliably.

It is common for the laptops used on the ISS (and formerly the shuttle) to lock up or have data corruption. They aren't used for mission and safety critical systems so such behavior is tolerable. That won't be the case on a satellite intended for more than a gee whiz tech demo.

Many satellites suspend operations when passing near the South Atlantic Anomaly because of the disruption is causes. That is with purpose built, space qualified hardware.

Re:Sort of pointless (0)

Anonymous Coward | about a year ago | (#43112665)

And NASA'a bitchin about under funding ,you have a space agency already testing this, but NASA wants to waste money to do this same experiment themselves?

Re:Sort of pointless (1)

FalconZero (607567) | about a year ago | (#43104795)

I'm tempted to agree. I don't really understand the point of this shoehorning - apart from to inspire.

If the article is to be believed, and the phone is completely unmodified, I straight away see a number of issues :
  • * The battery will be way outside it's operating norms - likely to alternately produce very little power, and explode due to overheating.
  • * The electronics is specifically designed to be small, and consequently more vulnerable to radiation.
  • * Half of the mass of the device is unnecessary. For example, there is no point having a touchscreen display if nobody is there to look at it or touch it.
  • * The unmodified operating system is completely geared towards having a user interacting with it. There are far more suitable (free) alternatives available off the shelf.

Re:Sort of pointless (1)

h4rr4r (612664) | about a year ago | (#43104981)

Why would it need its battery? It can be removed to save mass and powered by the satellite. If not it can sit there and do nothing.

Radiation below the van allen belts is not that bad. Works fine.

If you are worried about the mass of the display remove it. Not exactly hard to do.

Modifying android is simple, just download the sources and make your changes. Hell you can have the thing just boot linux without the android layers if you want.

Re:Sort of pointless (2)

squiggleslash (241428) | about a year ago | (#43105149)

But if it doesn't need a battery what's the advantage of this anyway? I mean, the Nexus seems to have a lot of crap (screen, speakers, cameras - well, I guess they may be useful, GSM/UMTS radio, bluetooth radio, Wi-fi radio, GPS, and probably a crapload more I can't think of at the moment) that seems completely useless for a device intended to control a satellite, while being missing almost everything needed aside from the CPU, memory, and storage.

You know, if only there was some kind of cheap computing device, much much cheaper than a smartphone, that didn't have all that unncessary stuff, maybe something with slightly more in the way of IO features, albeit not enough for this application, maybe with a name that's the combination of a fruit and a mathematical symbol and constant so you know it's going to be high quality?

Re:Sort of pointless (1)

h4rr4r (612664) | about a year ago | (#43105287)

Which has a CPU so many times slower.

I like the RaspberryPi as much as many others, but where can I get faster version of it? As cheap as an old used smartphone. Which would be $100 tops.

Plus by using a Nexus One you can likely get more publicity and quite possibly sponsorship.

Re:Sort of pointless (1)

squiggleslash (241428) | about a year ago | (#43105827)

I was actually talking about the Strawberry e, but WHATEVER.

More seriously, I'm inclined to think that the processing power needed for this application means that both the Raspberry Pi and the Nexus One are more than capable of doing it. Indeed, I suspect an old Nokia N800 would have enough power. Or an iPod.

I can't help but feel that in a context where you need very little processing power, but a lot of reliability, a smartphone is a terribly bad idea.

GPS (1)

DrYak (748999) | about a year ago | (#43112679)

GPS doesn't work [wikipedia.org] (by design) at this altitude.

If you're high than a set altitude and (18km) move faster than a threshold speed (515m/s) (which a satellite qualifies for both) the GPS chips refuses to give accurate readings (by design, so it can't be used to build cruise missiles and similar).

(although at this altitude in theory you should be able to get signal from much more GPS satellites with less atmospheric distortions, and thus get a better reading. Also, a satellite move in a much more regular fashion, so it should be easier to compensate for the algorithmic lag. So accuracy could be good, but rules prevent it).

Well maybe some foreign phones (which were never meant to step on US territory) might use ITAR-free GPS chips.

Beside GPS, thermal, compass and camera are other sensors available on a smartphone which might be useful.

Re:Sort of pointless (1)

FalconZero (607567) | about a year ago | (#43105257)

But that's the point - they're making a point out of the fact that it's unmodified.

If they're willing to modify it, then ok - lets throw out the screen, the battery, and the speakers. And since we're doing that, why don't we modify/remove the chassis - as it's primary design consideration is the parts we're stripping out.

And yes, we can replace the O/S with a modified, stripped down, or completely alternate one....

Every step they take in adapting a phone to better suit the operating environment is a step towards other existing off the shelf solutions.

Re:Sort of pointless (3, Insightful)

Bill_the_Engineer (772575) | about a year ago | (#43105317)

The article mentions that the battery is still installed in the phone. Keep in mind that this is a cubesat and things like temperature control, single event upsets, and outgassing are usually not considered mission critical since these are designed to be short lived amateur satellites in LEO (these sometimes live longer than planned, just no extra expense was made to insure longevity). This is to keep costs low while maintaining the programs original intent of education.

Re:Sort of pointless (1)

Bill_the_Engineer (772575) | about a year ago | (#43104823)

Off-the-shelf hardware, though - that would be a bigger deal. It's doomed to failure, but if somehow it could work that would be huge.

This is not the first satellite that used much cheaper off-the-shelf hardware. They have already proven successful in their LEO missions.

Re:Sort of pointless (2)

alen (225700) | about a year ago | (#43104835)

you can customize android for your environment. download it from AOSP and add any drivers you need, etc

iOS, Win Phone and others are products you buy as is and use as is. most you can do with iOS is buy the enterprise software license to load your own apps outside the app store

Re:Sort of pointless (2)

PoolOfThought (1492445) | about a year ago | (#43104855)

Exactly! The OS of a smart thermostat (5-10 years ago) might be good enough to control a "lowcost space craft". Can this OS context switch? Yes. Can it handle IO to different channels? Yes. Will it stay up and running assuming no one does something stupid and crashes it? Yes! Woo hoo we have a winner for an OS.

The hardware is the key. If the hardware can survive space and operate as expected then you've got something. Even if it doesn't operate as expected, but it does operate consistently you MIGHT be able to work with that. If the hardware works and the OS quits just because it's in space then the OS probably isn't fit for operation anywhere.

Re:Sort of pointless (1)

viperidaenz (2515578) | about a year ago | (#43111705)

Just comment out the following code:

if (in_space) {
        CRASH_OS();
}

Re:Sort of pointless (1)

bWareiWare.co.uk (660144) | about a year ago | (#43104945)

Costs aren't only the license fees paid (and this is operating outside any copyright territory so licences fees would be extremely prohibitive to enforce anyway).

The biggest cost of Windows/OS:X is that can't make changes. A satellite may well have hard real-time requirements or require other kernel changes that exist for Linux but not for closed source general purpose OSs.

Re:Sort of pointless (1)

BitZtream (692029) | about a year ago | (#43107873)

Microsoft will sell you what ever you want for the right price. Plenty of people have the source code to Windows's kernel already.

Re:Sort of pointless (2)

h4rr4r (612664) | about a year ago | (#43104947)

The cost is not the issue.

With android modifying the OS is possible and since it runs linux very well known by the community at large. You can build your own custom version of android and boot that. This is not so with the other options you mention. Using a smartphone is likely cheaper than other off the shelf options since they are sold in such high numbers.

The real cost they are controlling for is mass. Getting that mass to orbit is the most expensive part of this whole thing.

Re:Sort of pointless (1)

93 Escort Wagon (326346) | about a year ago | (#43107889)

With android modifying the OS is possible and since it runs linux very well known by the community at large.

If you're going to modify the OS, Android is superfluous - just go with straight Linux or one of the BSDs.

But how will they ... (0)

Anonymous Coward | about a year ago | (#43104711)

... Yank the battery whin it inevitably freezes up?

Can you hear me now? (1)

nomorecwrd (1193329) | about a year ago | (#43104725)

... and now?

Now?

real links (0)

Anonymous Coward | about a year ago | (#43104787)

Mission Page [sstl.co.uk]

and

FAQ [sstl.co.uk]

I was curious about radiation and temperature affecting a phone that is "bolted to the bottom" but it appears to be more TFS lies. In the FAQs, they say that the phone is against a panel looking through a "port hole" and that they have taken radiation into account which means it is shielded somehow, even though they don't say how exactly.

Why not use Raspberry Pi? UK innovation. (1)

patiwat (126496) | about a year ago | (#43104871)

If Surrey was actually trying to promote UK technology, they'd have used a Raspberry Pi :P

Re:Why not use Raspberry Pi? UK innovation. (3, Informative)

joh (27088) | about a year ago | (#43104965)

This mission was fixed years ago, there was no Raspberry Pi back then.

Re:Why not use Raspberry Pi? UK innovation. (1)

Bearhouse (1034238) | about a year ago | (#43104983)

Ah, but it's not an "Android phone", hence not cool enough.
Also, one would have to admit that the Pi has had a few issues...maybe not ready for space yet.
Otherwise, I completely agree with you; it's got about the same processing capacity, plenty of distro choices, good dev support, nice inbuilt video...

Re:Why not use Raspberry Pi? UK innovation. (3, Informative)

Eggbloke (1698408) | about a year ago | (#43105293)

I was at a talk from one of these guys once and from what I remember he said that the amount of things you get with a phone make it attractive. They have a camera, temperature sensor and compass straight off as well as probably more stuff. A Raspberry Pi probably would be better with some more work though.

Android is cool, but is not a RTOS (1)

Bearhouse (1034238) | about a year ago | (#43104895)

I like the idea of simply taking a smartphone and sticking in the box. OK, sending up the case and display is maybe a waste...but on the other hand, a bit less systems integration work to do.

Shame 'the artist formally known as RIM' took QNX back to closed source; that's a really great RTOS.
C'mon guys, publicity like this would help you get some 'buzz' back.

I love this part especially: (4, Interesting)

QilessQi (2044624) | about a year ago | (#43104967)

"The onboard computer will monitor the temperature of the phone battery. If it sees it is getting too cold, it will trigger a processor intensive program to run on the mobile phone, which will warm it up."

Next time I'm out on a winter day, I'll just turn on my Live Wallpaper with Conway's Life running on an infinite grid. Instant pocket hard-warmer!

Re:I love this part especially: (2)

DigiShaman (671371) | about a year ago | (#43105413)

Be careful! Some phones have been documented to burn a hole through the pocket. Literally. Though if I had to guess, the owners of those phones were probably using an aftermarket pirated battery of dubious quality.

Re:I love this part especially: (0)

Anonymous Coward | about a year ago | (#43106231)

What the fuck is a pirated battery?

Re:I love this part especially: (1)

DigiShaman (671371) | about a year ago | (#43106885)

A pirated battery is one in which the battery itself is made in an unknown origin and labeled with counterfeit branding/labeling. So those $1.25 new Motorola Droid 2 batteries on Amazon may in fact be a bogus. Just lookup "Chinese Fakes" in the reviews for more information. Buyer beware.

Re:I love this part especially: (0)

Anonymous Coward | about a year ago | (#43107609)

so its not pirated at all then, its counterift... do you perchance work for the riaa?

Temperature? (1)

Dunge (922521) | about a year ago | (#43105137)

Pretty sure a normal Android phone don't resist outer-space temperatures.

Re:Temperature? (0)

Anonymous Coward | about a year ago | (#43105941)

There is no air in space. You'd actually have trouble keeping stuff cool.

Re:Temperature? (0)

Anonymous Coward | about a year ago | (#43107161)

I've designed COTS based power systems for Cubesat while working at Clyde Space and I think a normal Android phone may well be alright.

LCD screen in orbit? (1)

Coward Anonymous (110649) | about a year ago | (#43105215)

Phones have heavy touch screen LCDs (and other bits and pieces, like the case!) that are pointless in orbit. Did they really waste that much of their mass budget on an LCD touch screen? Or is the "stock nexus" on this thing really not so stock?

Re:LCD screen in orbit? (0)

Anonymous Coward | about a year ago | (#43105995)

Well they attached an industrial robot to push the buttons on the screen, naturally. How else would you control a phone?

Re:LCD screen in orbit? (1)

ThePeices (635180) | about a year ago | (#43107597)

Phones have heavy touch screen LCDs (and other bits and pieces, like the case!) that are pointless in orbit. Did they really waste that much of their mass budget on an LCD touch screen? Or is the "stock nexus" on this thing really not so stock?

I think that the rocket launching this into orbit has enough fuel and thrust to handle an additional 50 grams of mass.

Pics? (1)

mwn3d (2750695) | about a year ago | (#43105219)

Is it positioned to be able to take any pictures form space? They can probably turn the flash off. I don't think the earth is close enough for it to matter.

Radiation hardened? (1)

hawguy (1600213) | about a year ago | (#43105291)

I thought spacecraft used absurdly expensive radiation hardened 20 year old processors because providing enough shielding to prevent radiation from disrupting a conventional processor is weight prohibitive. Does this only apply to deep space probes?

Re:Radiation hardened? (0)

Anonymous Coward | about a year ago | (#43107129)

Not only deep space, but GEO as well. Military, industrial and automotive grade electronics are mostly fine for LEO, though. SSTL (the STRaND guys I think) has a lot of experience using them for LEO. I worked at SSTL for a couple years and did some work for them as a contractor years later. It's a fantastic company to work at / for.

Go Android! (3)

Graydyn Young (2835695) | about a year ago | (#43105435)

Allow me to be the first Android fanboy to say, "Suck it, IOS fanboys!"

Re:Go Android! (1)

BitZtream (692029) | about a year ago | (#43107931)

Seriously? Thats all you got?

http://www.youtube.com/watch?v=wtXquYhY7wo [youtube.com]

Define space, argue over that for a few days amongst yourselves, then get back to me.

Not that either one are particularly impressive feats as can be seen by the fact that anyone with the money can Google how to do it with either device, then pay a little bit of cash and well ... do it.

Its not like either one is doing anything that NASA didn't figure out how to do 60 years ago now.

Why Android? (1)

ogdenk (712300) | about a year ago | (#43106199)

Why Android, wouldn't a slim straight embedded GNU/Linux OS be a better choice from a reliability standpoint? Is there a robot finger for poking the screen? If not, Android in this situation was a solution in search of a problem.

Linux, QNX, FreeBSD or NetBSD would have all been adequate choices, likely more reliable and will all run on just about anything.

Re:Why Android? (0)

Anonymous Coward | about a year ago | (#43107151)

They wanted to figure out what to do with this old Nexus One now that they have a new Nexus4. It's obsolete hardware, sure, but it seems like a shame to throw it out when it still works as well as it did when new...

Re:Why Android? (1)

c0lo (1497653) | about a year ago | (#43107731)

It's obsolete hardware, sure, but it seems like a shame to throw it out when it still works as well as it did when new...

So, throwing it on a LEO doesn't equate to throwing it out? ;)

Re:Why Android? (1)

lister king of smeg (2481612) | about a year ago | (#43108857)

well android is linux under the hood, the point here is the common consumer hardware/software can do this unmodafied without much work

Re:Why Android? (1)

ogdenk (712300) | about a year ago | (#43109171)

After you dig through a bastardized Java VM. Why not just use the kernel and slap an embedded linux userland on it and call it a day? Would be much more suited to the task and make troubleshooting easier.

Low-Hanging Fruit (1)

ThatsNotPudding (1045640) | about a year ago | (#43107061)

Can you imagine the roaming charges!
Thank you! I'll be here all week!

and for that, I'm truly sorry.

Similar University Project (0)

Anonymous Coward | about a year ago | (#43107439)

I'm a part of a team at my uni that is doing something similar (tho not nearly as cool). We're launching a rocket for an April NASA competition with a scientific payload based around an older Android phone (+ attached IOIO). The phone is used to collect and analyze data and provide communication via SMS using a homebrew/compressed data serialization standard, but the recovery system uses a separate uC per competition guidelines. Otherwise, the whole thing would've been run of the phone. Android has proven to be a very elegant solution to our problem, and although we're not stress-testing the phone to space-like conditions, it has held up perfectly to our absurd launch impulses. In case you're interested in learning more, there's brief descriptions of the project on our site and lots of technical docs, too: http://uoflusli.com/ [uoflusli.com]

Requires a new Cell Phone Plan... (0)

Anonymous Coward | about a year ago | (#43108385)

[In Network]
Galaxtic Roaming: $1000/h

[Out of Network]
Galaxtic Roaming: $5000/h

Catch this (1)

ukoda (537183) | about a year ago | (#43113953)

I just hope they did proper thermal cycle testing before deployment. My Nexus One had a thermal related fault from the date of purchase that rendered it useless over time and HTC didn't want support it. Now it's just a pretty brick.

If their's fails too it could make for a funny support call:
Support: "Ok so you need to send it to our service center"
Owner: "Sure, just give us your Lat/Lon and we will de-orbit it over you, can you have someone go outside to catch it?"
Load More Comments
Slashdot Account

Need an Account?

Forgot your password?

Don't worry, we never post anything without your permission.

Submission Text Formatting Tips

We support a small subset of HTML, namely these tags:

  • b
  • i
  • p
  • br
  • a
  • ol
  • ul
  • li
  • dl
  • dt
  • dd
  • em
  • strong
  • tt
  • blockquote
  • div
  • quote
  • ecode

"ecode" can be used for code snippets, for example:

<ecode>    while(1) { do_something(); } </ecode>
Create a Slashdot Account

Loading...