Slashdot Mirror
Linux Rocket Blasts Off This Fall
HardcoreGamer writes "An Oregon amateur rocket group, the Portland State Aerospace Society, plans to launch a Linux-powered rocket weighing 12 pounds to 55,000 feet at a speed of Mach 3 in September, Wired News reports. The rocket's onboard computer is an AMD 586 processor and a Jumptec MOPS/520 PC/104+ board along with a power supply, a PCMCIA card carrier for an 802.11b card to transmit data to the ground, and a carrier board for a 128-MB CompactFlash card for long-term storage. The flight computer runs a stripped-down version of Debian Linux, with the 2.4.20 Linux kernel. The group will present a paper (HTML | PDF ) on the use of free software in rocketry at Usenix 2003. The real question is whether their network card will survive 10 seconds at 15 Gs!"
I guess this redefines the term "crash."
"Men lie."
"Yeah, about sleeping with other women, but never about bioluminescent plankton."
-Dan Brown
Do you really want to have to pay royalties to SCO on your rocket? There are high-quality commercial embedded OS's without much clearly defined IP rights, and no such liability issues, and I think its a good idea to go with the Gartner recommendation and avoid the potential legal issues with Linux for the time being.
Imagine a Beo- aah, forget it.... :-)
Huxley
PS please don't hurt me...
and of course this will just encourage those rascally terrorist who want to build nasty rocketses and blow us all to smithereens. Since now they won't have to pay those pesky licence fees for operating systems for their WMDs.
802.11b for data transmission to the ground? I know my 802.11b network doesn't have a range of 55,000 feet.
I think the real question is will the pringles can survive 15 g's for 10 seconds?
I browse Slashdot at +3, Funny
"...MS-built engine"
It'll never get off the ground - too much bloat!
Go permanent? In your dreams and my worst nightmares.
But, I gather the greatest stresses will be on the computer hardware, as 10 G's will put a meaningful load on the parts, not to mention vibrational loads. And rockets are difficult to begin with. Here's hoping it works.
Bush: linux can be use to launch rocket? The very thing that terrorist lacks? It's free and distributed widely on the Internet? We got a problem here Ashcroft: not only that, but its source code is not encrypted, anyone could store a copy in their compueter. Bush: Then I'm assuming that even if we EMP all the computers, the source might still be stored somewhere as a printed copy? Ashcroft: I'm afraid so. I always have a problem tainting uses of technology Bush: then let's ban printers as well, that will buy you sometime.
Ours is bigger.
--Just the place for a snark!
Heck, I'd worry about the CF card. I doubt it's a hard disk (of the spinny-type) as the paper states, as that'd crash on either liftoff or chute deployment. I'd bet it's a flash-type, just like a simple camera memory card. And then I'd wonder whether it'd survive too. Many of them have altitude restrictions (though I seriously doubt they're for real - it's probably a "don't use this in an airline design!" warning) as well. Remember to put some sort of retaining mechanism on the CF slot. Wouldn't want the card pulling out on liftoff, now would you. :)
Yipes. High-altitude, high stress stuff is always a pain (which is why aerospace companies make so much money designing things).
It'll definitely be cool to see if this works. The paper's a little light on details of the design (for certain things - like the actual construction or parts choices - for other things it seems pretty detailed).
10g is not a lot of acceleration for electronics, as long as large components are securely fastened (even with tie-wraps), and there are no moving parts.
The only non-solid-state parts on the design are the connectors, which can handle hundreds or thousands of g's of acceleration without "bouncing" on the pins.
PC-104 is designed for high-stress applications such as this.
...that my design documents aren't the only ones that look like this.
MS are great. Linux is crap.
Now simply ride the flames that come out the back of that.....
Aim for Redmond, guys.
Bowie J. Poag
I would expect that the guidance systems for most rockets built in recent years would use a real-time OS like QNX or Chorus, not Linux.
Never eat more than you can lift -- Miss Piggy
For the longest time the software was written by hand from the metal up. You can't afford to have one bug in space code. It could cost half a billion dollars. Every routine was coded three different ways, and three systems ran separately. If they ever disagreed you knew you had a problem. So while you have an OS of sorts, it's the Shuttle OS, and nothing else. After all, there are a thousand assumptions that OS developers make that a space programmer has to choose him or herself. In Linux, the coder is always saying "this amount of precision is ok," but for a rocket the amount of precision needed is very well known, and incredibly demanding at all levels. For a hobbyist group, linux is one thing, but if you want to put something in geosynchronous orbit indefinitely...
As others have pointed out, it is not *linux* powered. But now thanks to Ashcroft and his straight man, bin Laden - anyone using model rocket fuel is considered a terrorist threat. So, not only is not linux powered, it probably won't be powered at all.
When information is power, privacy is freedom.
But as previously posted, what OS has been guiding rockets? Wind-SCO-s?
Uhm, nope. You know there _are_ other OSs out there besides Windows and Linux. Sure, a post like that makes for good Karma, but for crying out loud, what is the exact value of your post?!
I will tell you which OSs have been running stuff like this; real-time, embedded OSs, such as VxWorks, QNX and all the others. Until recently linux SUCKED ASS for real-time applications. I don't think even Torvalds would mind me saying so. It just wasn't designed that way. There have been major improvements lately, which are all very promissing, but for applications that really demand real-timeness, probably very few people (in their right frame of mind, no offense) would choose Linux.
Sure the OS may be a version of Linux, but the really interesting part is that they've found a way to harness all of that heat from the AMD to get the rocket that far up!
:)
Talk about potential for burn up on reentry though.
Not only did Linux "Suck Ass" for doing hard realtime, most of the architectures that it runs on do as well. For tasks of this nature, unpredictabilities like cache misses can be deadly, so you want much of the critical control features to run on as simple an architecture as possible.
PIC or MC68hc11 are good candidates, anything more complex than an 8080 probably isn't.
If it's controlling thrust vectoring, control surfaces, or fuel valves, I sure wouldn't want an OS like Linux, Windows, HPUX, Solaris, etc with multitasking and/or VM.
I was about to ask what the specific impulse of Linux was and if it changed from Intel to AMD to PowerPC. And if NASA knows about it. Heh...Imagine a Beowulf BOOSTER of those.
If Mr. Edison had thought smarter he wouldn't sweat as much. --Nikola Tesla
Aww c'mon, they've only Linux. Not as if installing Linux is rocket-science...
More than mere navel gazing.
During system installation, it's important to use the right networking packages, to cope with the slightly nonstandard hardware. At the bash prompt, type:
% apt-get skynet
http://cvs.psas.pdx.edu - Have fun!
... have been doing this for awhile. The PC104 stack in their VTVL rockets/crafts have always been linux kernels.
He's also been using 802.11 for communications.
His laptop control station is win32 though.
ArmadilloAerospace
-malakai
-Malakai
A Dragon Lives in my Garage
You should have seen what I did to my network card after it stopped working! Amazingly, after an approximate 20G throw against the wall, it started working again!
Get rid of everything Micro and Soft: Buy Viagra and/or Linux
Comment removed based on user account deletion
Oh the shame, caught in the act. I'm guilty of snide remarks about a topic that's actually important.
:) - want a pretty good amateur avionics system that eventually will be capable of active guidance? Heck, you could lose the fins and get another few thousands feet.
Let me extend your comment with a concern of ours: the lack of technical collaboration between amateur aerospace groups.
When we started PSAS, there was very little posted on advanced amateur rocketry (especially avionics). Now there's a lot more, but for some reason many amateur aerospace groups either:
1) feel their technology is good enough to be proprietry (i.e., closed source), or
2) or they're too lazy to post their results.
Either way, the community loses and we have to slog through all the stupid mistakes in order to recreate what they've done.
This is incredibly frustrating - so when we started PSAS one of our goals was to always post everything we learned. We're open source, open hardware, and more importantly, open to the community: we're trying to collaborate with as many amateur groups as we can (in fact we've currently got collaborations going on with two other universities).
In fact, you should be able to recreate _everything_ we've done by following our history and getting the technical details (schematics, firmware, software, system diagrams) from our site.
So, to the BYU people with a _much_ bigger rocket
Contact us, let's collaborate.
(and it is funny, by the way), but this attitude is not far from the truth. I happen to be a member of Tripoli (one of the associations that governs high power rocketry) and we currently have a letter writing campaign going on. The government is trying to make it harder for us hobbyist to get engines over certain power ratings. I guess they fear someone using these engines to power guided rockets. This is kinda silly, as most terrorist have access to better weaponry, ie Stinger missles. Of course, the gov is also afraid that the terrorist do all there planning on the internet, too.
"...MS-built engine"
It'll never get off the ground - too much bloat! "
"...Linux-built engine"
It'll never get off the ground - nobody wrote the man page for the launch command!
"Derp de derp."
modprobe: Can't locate module podbaydoors
They also have to comply with U.S. export laws (International Traffic in Arms Regulations, 22 CFR 120-130) that require an export license for missile technology. That trumps the GPL. You can get in serious trouble if you ignore the law.
Mea navis aericumbens anguillis abundat
First of all, the techs will spend 3 weeks just trying to install Linux. There won't be a single driver that's compatible, and the few that exist will be buggy. Each different tech will want a different version, one wants NASA-Linux, another wants Goddard-Linux, and they all will be uninstalling the previous install and secretly putting their own distibution on it. If they ever settle on one install, then they will discover there's no applications to run, except Windows versions. Finally they'll get fed up with it and just put OSX on.
What would RIAA do if a satellite full of mp3s was launched with easy access for everybody?
Easy, sue SCO, the owner of the patents for Linux for developing a platform that allowed such a satellite to be launched.
Just imagine a MIRV cluster of these!
Wah!
I think a bigger concern would be whether the connectors are properly held together and maintain electrical connection. The boards should be fine.
You can find lots of DEBI info by looking through the past two weeks of my journal. You'd have to follow links from my web page link below in my sig. I won't link it directly since the machine will probably tank after only a few concurrent connections.
I have to wonder if 802.11b is really all that great of an idea for this. A person could build a jammer for $5 in radio shack parts that would crash this thing.
Not sure if they tested for this but if they didn't I think this particular rocket might not go too far.
"This is certainly a brave approach that throws everything we thought we knew about building a rocket" said NASA Ames' deputy director for research, G. Allen Flynt. "It shows that we've being doing it all wrong for years, trying to build ever more powerful, more efficient rocket motors, when the real solution was staring us in the face; Replace the expensive rocket motor with a cheap commodity PC running GNU/Linux. Brilliant. My hat goes off to these guys"
You can't win Darth. If you mod me down, I shall become more powerful than you could possibly imagine
How much will travelling at Mach 3 affect the frequency of the 802.11b signal received at ground level?
15G is nothing - that's like dropping the card onto carpet from about 2ft. Not exactly stressful for some solid state hardware - even a hard drive could probably cope with that while running. The duration doesn't make much difference - providing they don't exceed the amount of G required to break something (probably more like 80+G). The vibrations might cause the G level to peak much higher than the overall accelleration of the rocket however.
I would have thought that vibrations are much worse than the overall acceleration of the rocket; Anyone ever taken a computer out of the back of a car (which probably never exceeds 1.5G) only to find that some screws have come loose or a PCI card has fallen out? (cos I have!).
NEWS JUST IN---->The RIAA has sued the makers of said rocket, as the 802.11b link could "techinically be used to share illegal files accross the network".
Since we're talking about rocketry, I think we should be using sensible units, not these so-called "feet".
FYI: 55000 feet are 543 femtoparsecs, or 1.77 picolightyears, or 112 nanoastronomicalunits.
Or 16.75 kilometres, while we're at it.
I though the first linux powered rocket took flight 2 years ago...
I remember they used the jumptec 386 dimmpc and used ham radio packet on 144/440mhz to get telemetry up/down.
I know I saw it here. Can anyone find it?
Do not look at laser with remaining good eye.
Paint on the side... "Try and examine the code on this SCO" :)
Linux really is rocket science.
Oh, and does anyone know the GPS coordinates of either SCO-HQ or redmond?
:)
Or both if we are really going cluster
I for one, welcome our new hot grits... PROFIT!
I can see my /home from here!
PC104 is NOT a good bus for hard real time applications. It hasn't got enough bandwidth or interrupt support. Couldn't they at least go PCI or CPCI?
Maybe it's not so bad. It sounds like all the mission critical I/O is on the CPU board, so it doesn't go through that damn bus. But even so, it makes me wonder about their thought processes. Maybe it'll work. But if it does, it won't be because of PC104. It'll be despite it.
PC104+ is a stupid kluge. Adapting an architecture from the IBM PC-AT to a class of applications it was never intended for. PC104+ is to hardware what Windows CE is to operating systems.
Oh well. At least they're not using Windows CE.
http://www.angelfire.com/ca3/marlowe Better a smartass than a dumbass.
Finally, somebody gets an AMD to run at a high speed.