| User | Post |
|
6:05 am January 14, 2010
| Luke Maurits
| | Adelaide, Australia | |
| Admin
| posts 1483 | |
|
|
Stumbled across this today: The BeagleBoard is a small, fanless, USB powered single board computer. It's about 3 inches square, costs $150 and has a DSP unit onboard which is powerful enough to encode HD video. The schematics and other design materials are available online, although I'm not sure what the license is – at least the board is "semi open".
Would something like this be suitable for use as our computer? It is very affordable and also quite small which means that we can fit several of them onboard for redundancy, and that we can build a radiation shield for them which is fairly small (and hence light).
|
Main CLLARE workgroups: Mission Planning, Navigation and Guidance. I do maths, physics, C, Python and Java.
|
|
|
9:34 pm January 14, 2010
| Rocket-To-The-Moon
| | Altus, Oklahoma, USA | |
| Member | posts 685 | |
|
|
Post edited 3:36 am – January 15, 2010 by Rocket-To-The-Moon
Although I can't say this is our best option, it is definitely along the correct line of thought. Is there any inherent advantage of something like this over a complete package like a netbook computer?
These are small and cheap enough that we could easily run several of them in parallel for redundancy.
|
Main Workgroups: Propulsion & Spacecraft Engineering
|
|
|
9:45 pm January 14, 2010
| Luke Maurits
| | Adelaide, Australia | |
| Admin
| posts 1483 | |
|
|
- The BeagleBoard is a 3 inch square. That's far smaller than the smallest notebook you're likely to find.
- With notebooks you'd need to perform a little surgery to remove monitors which would be of little use
I think the main advantage of this over a netbook would be size. The BeagleBoard is a 3 inch square, even the smallest netbooks have 7 inch screens which makes them a lot bigger than this. However, there's also cost. When you buy a notebook you are paying not just for a mobo, CPU and RAM, you're paying for a monitor, battery, and other stuff we don't need. US$150 won't buy you any netbook that I'm aware of. Also, netbooks come with monitors, keyboards, etc. that you would need to carefully remove before including them in the CM, where as BeagleBoards could just be mounted in our own cases.
While the BeagleBoard may not be the best product out there for our purposes, I think that something like the BeagleBoard would be an awful lot more appropriate than a netbook.
A thought: The BeagleBoard seems to have a fairly good graphics capabilities for its size, which we won't really be needing. We could look into using the GPUs as a source of extra computational grunt. This is a fairly trendy area of research these days, I bet we could find resources to make it work.
|
Main CLLARE workgroups: Mission Planning, Navigation and Guidance. I do maths, physics, C, Python and Java.
|
|
|
9:48 pm January 14, 2010
| Rocket-To-The-Moon
| | Altus, Oklahoma, USA | |
| Member | posts 685 | |
|
|
Post edited 3:49 am – January 15, 2010 by Rocket-To-The-Moon
I guess I was considering using the netbook's screen and input devices as the primary human interface to the spacecraft.
Maybe a netbook as a frontend and then these BeagleBoards running the mission software?
|
Main Workgroups: Propulsion & Spacecraft Engineering
|
|
|
9:54 pm January 14, 2010
| Luke Maurits
| | Adelaide, Australia | |
| Admin
| posts 1483 | |
|
|
I had actually always imagined us having multiple screens, although I've never said this anywhere. 7-9 inch LCDs are so cheap because they are mass produced for netbooks and portable DVD players that we could easily afford to have 4 or 6 of them on the instrumentation panel, with each screen showing a separate set of related data, e.g. a navigation screen, a cabin sensor screen, etc. Each physical screen could have a button underneath it to cycle it through the various "virtual screens" involved. That way if a screen dies mid-mission, one could just cycle them around so that the least criticals virtual screen at any moment was the one not being shown.
I haven't thought much about actual input. A netbook keyboard might be too small and fiddly, especially in gloves. Also, hopefully the running of the CM shouldn't involve so much input that a complete keyboard is necessary. Maybe touch screens would be a good idea?
|
Main CLLARE workgroups: Mission Planning, Navigation and Guidance. I do maths, physics, C, Python and Java.
|
|
|
7:21 pm January 15, 2010
| brmj
| | Rochester, New York, United States | |
| Member | posts 402 | |
|
|
The beagle board looks allright, but perhaps hot ideal. It has more graphical capabilities then we would need, for one thing.
I think either netbook guts or any number of readily available embedded GNU/Linux devices would work just fine for our computers. I think the idea of using GPU computing is probably a bad one, since the one thing it is really, really good at is massively parallel computation. For a few threads, it's better to stick with CPUs.
IO devices: I agree that netbook keyboards are too small for gloves. Perhaps one of those infernal roll-up keyboards with the keys way the hell apart. They are very durable, and though they are terrible to type on, it's not like our astronaut will be writing a novel. I think touchscreens are bad news, since they can be finicky and they don't lend themselves to complex and adaptable interfaces. We definitely need to have a full keyboard, at least in case something goes wrong and a more adaptable interface than point and click is required. For screens, I agree that several might be nice. What I am picturing is between one and four small LCDs cycling through various instrument readouts, and a somewhat larger main screen with a full, multipurpose but highly customized GNU/Linux environment with virtual desktops.
|
Main work groups: Propulsion (booster), Spacecraft Engineering, Computer Systems, Navigation and Guidance (software)
|
|