Author Topic: Arduinos and Raspberry PIs at ISS  (Read 25310 times)

Offline Robotbeat

  • Senior Member
  • *****
  • Posts: 39270
  • Minnesota
  • Liked: 25240
  • Likes Given: 12115
Re: Arduinos and Raspberry PIs at ISS
« Reply #20 on: 02/26/2015 04:34 pm »
Arduino is just a certain environment for a microcontroller. But if you can program C, no reason you can't use another microcontroller, and several microcontrollers have been used in LEO.

There's a big community around microcontrollers in general of which Arduino is a part.

But I believe the Arduino environment CAN compile for other microcontrollers which have more experience in orbit.
« Last Edit: 02/26/2015 04:45 pm by Robotbeat »
Chris  Whoever loves correction loves knowledge, but he who hates reproof is stupid.

To the maximum extent practicable, the Federal Government shall plan missions to accommodate the space transportation services capabilities of United States commercial providers. US law http://goo.gl/YZYNt0

Offline Robotbeat

  • Senior Member
  • *****
  • Posts: 39270
  • Minnesota
  • Liked: 25240
  • Likes Given: 12115
Re: Arduinos and Raspberry PIs at ISS
« Reply #21 on: 02/26/2015 04:43 pm »
...
- The risk of an event depends on the silicon size not packaging, so for the same amount of flash at the same process geometry, the risk is roughly the same....
Right, and the amount of flash on an Arduino is tiny (256kB on the Megas and smaller on the regular Arduinos), 4 orders of magnitude less than on the Raspberry Pi (4GB is the minimum recommended). And the Raspberry Pi generally NEEDS that much flash just to boot up the Linux environment (you might gain almost an order of magnitude by using a compact Linux distribution), so this is a valid comparison (provided Arduino is sufficient). If flash is the weak link, then Arduino is lightyears ahead.
« Last Edit: 02/26/2015 04:43 pm by Robotbeat »
Chris  Whoever loves correction loves knowledge, but he who hates reproof is stupid.

To the maximum extent practicable, the Federal Government shall plan missions to accommodate the space transportation services capabilities of United States commercial providers. US law http://goo.gl/YZYNt0

Offline A_M_Swallow

  • Elite Veteran
  • Senior Member
  • *****
  • Posts: 8906
  • South coast of England
  • Liked: 500
  • Likes Given: 223
Re: Arduinos and Raspberry PIs at ISS
« Reply #22 on: 02/27/2015 09:40 pm »
I am not looking for the lowest cost functionality.

I am looking for support of the most popular standardized systems. The goal is to open up ISS experimentation to the most users, not to the most sophisticated who can develop radiation tolerant systems the size of a fingernail.

My expectation is that a system designed today to support Arduinos and Raspberry PIs would  be still supporting those systems years from now, except that those systems will have greater capacity than today.

The company that designs the radiation tolerant system does not have to be the school boy experimenters. If designing and testing radiation tolerant system costs say $500,000 and the company expects to sell 100

$design cost / quantity = $cost overhead per chip.
$500,000 / 100 = $5000

Expensive for a school boy but plenty of laboratories can afford $5000 for a space certified controller

Offline Danderman

  • Extreme Veteran
  • Senior Member
  • *****
  • Posts: 10288
  • Liked: 699
  • Likes Given: 723
Re: Arduinos and Raspberry PIs at ISS
« Reply #23 on: 02/28/2015 04:36 pm »
No good.

A commercial system requires that experimenters be able to use out-of-the box Arduinos and Raspberry Pis.

Any system that requires an add-on cost for radiation tolerance in the nature of $5000 will fail commercially, although it may be a great technical success.

If you recall, Iridium and Globalstar failed commercially - partially because users could not use their own cell phones to use the system, they needed a special handset.


Offline Robotbeat

  • Senior Member
  • *****
  • Posts: 39270
  • Minnesota
  • Liked: 25240
  • Likes Given: 12115
Re: Arduinos and Raspberry PIs at ISS
« Reply #24 on: 02/28/2015 06:39 pm »
Iridium was reborn and has succeeded in the original subscriber goals. Not a perfect example.
Chris  Whoever loves correction loves knowledge, but he who hates reproof is stupid.

To the maximum extent practicable, the Federal Government shall plan missions to accommodate the space transportation services capabilities of United States commercial providers. US law http://goo.gl/YZYNt0

Offline Danderman

  • Extreme Veteran
  • Senior Member
  • *****
  • Posts: 10288
  • Liked: 699
  • Likes Given: 723
Re: Arduinos and Raspberry PIs at ISS
« Reply #25 on: 03/01/2015 03:57 am »
Iridium was reborn and has succeeded in the original subscriber goals. Not a perfect example.


Iridium was one of this country's worst business failures.

The point here is that there two paths to hosting Arduinos at ISS

a- require special versions of Arduinos that are more tolerant of radiation but will cost each user a boatload of cash

or

b- develop a host for Arduinos that would mitigate radiation issues and allow standard Arduinos to be used at ISS.

Both are technically possible, but only one would succeed.

Offline sdsds

  • Senior Member
  • *****
  • Posts: 7201
  • “With peace and hope for all mankind.”
  • Seattle
  • Liked: 2050
  • Likes Given: 1962
Re: Arduinos and Raspberry PIs at ISS
« Reply #26 on: 03/01/2015 04:37 am »
In the worst case (total Arduino lockup/lockout) I think you would want your host to have a connection to the JTAG header on the Arduino. That should get the Arduino going again, unless it suffered an error that was only correctable with a soldering iron!

https://en.wikipedia.org/wiki/Joint_Test_Action_Group#Storing_firmware
— 𝐬𝐝𝐒𝐝𝐬 —

Offline A_M_Swallow

  • Elite Veteran
  • Senior Member
  • *****
  • Posts: 8906
  • South coast of England
  • Liked: 500
  • Likes Given: 223
Re: Arduinos and Raspberry PIs at ISS
« Reply #27 on: 03/01/2015 07:49 pm »
Iridium was reborn and has succeeded in the original subscriber goals. Not a perfect example.


Iridium was one of this country's worst business failures.

The point here is that there two paths to hosting Arduinos at ISS

a- require special versions of Arduinos that are more tolerant of radiation but will cost each user a boatload of cash

or

b- develop a host for Arduinos that would mitigate radiation issues and allow standard Arduinos to be used at ISS.

Both are technically possible, but only one would succeed.


That depends whether the manufacture can find more customers for his computer module. R&D having been paid for the second batch will be considerable cheaper.

Offline Robotbeat

  • Senior Member
  • *****
  • Posts: 39270
  • Minnesota
  • Liked: 25240
  • Likes Given: 12115
Re: Arduinos and Raspberry PIs at ISS
« Reply #28 on: 03/01/2015 09:55 pm »
Iridium was reborn and has succeeded in the original subscriber goals. Not a perfect example.


Iridium was one of this country's worst business failures.

The point here is that there two paths to hosting Arduinos at ISS

a- require special versions of Arduinos that are more tolerant of radiation but will cost each user a boatload of cash

or

b- develop a host for Arduinos that would mitigate radiation issues and allow standard Arduinos to be used at ISS.

Both are technically possible, but only one would succeed.
You don't have an "A or B" choice. There are literally dozens of microcontrollers that are compatible with the Arduino environment, essentially all of them are cheap, and some have even flown into space before.
Chris  Whoever loves correction loves knowledge, but he who hates reproof is stupid.

To the maximum extent practicable, the Federal Government shall plan missions to accommodate the space transportation services capabilities of United States commercial providers. US law http://goo.gl/YZYNt0

Offline Danderman

  • Extreme Veteran
  • Senior Member
  • *****
  • Posts: 10288
  • Liked: 699
  • Likes Given: 723
Re: Arduinos and Raspberry PIs at ISS
« Reply #29 on: 03/02/2015 11:06 pm »

You don't have an "A or B" choice. There are literally dozens of microcontrollers that are compatible with the Arduino environment, essentially all of them are cheap, and some have even flown into space before.

OK. Let me try to explain this way:

In developing an ISS based system to host popular microcontrollers, the trick is to decide on hosting those microcontrollers with a large installed base of users, and then devise a system which allows those users to develop payload systems that require the least amount of modifications - ie make the customers do the least amount of work to use the system.  To extent that systems other than Arduinos and Raspberry Pis are physically compatible, then great their users can play, too.

This is kind of like the PC business in 1983, there were a bunch of quasi-compatible systems out there, some that ran MS-DOS but were not PCs, but most of them died pretty quickly, leaving just the true compatibles. I would expect a similar shakeout in this field.

A secondary concern is that although Arduinos and Raspberry Pis will likely be around in advanced - but compatible - forms for many years, some of the other systems may not be. 

The goal here is to be able to host as many users as possible for the lowest cost, so that the maximum number of space experiments are supported. I still have no clue how to implement this vision, and "space is hard", but I may give it a try.


« Last Edit: 03/02/2015 11:09 pm by Danderman »

Offline baldusi

  • Senior Member
  • *****
  • Posts: 8356
  • Buenos Aires, Argentina
  • Liked: 2539
  • Likes Given: 8273
Re: Arduinos and Raspberry PIs at ISS
« Reply #30 on: 03/03/2015 11:06 am »
If I may give you my economist view, what the expected cost for the client, for the infrastructure (i.e. mechanical, electronics, general integration, shipping, processing, launch, etc.). Do $35 vs $350 make a big difference? How much does 300g of shielding cost? How much does an exta 300cc of shielding cost?

Offline Danderman

  • Extreme Veteran
  • Senior Member
  • *****
  • Posts: 10288
  • Liked: 699
  • Likes Given: 723
Re: Arduinos and Raspberry PIs at ISS
« Reply #31 on: 03/03/2015 02:19 pm »
If I may give you my economist view, what the expected cost for the client, for the infrastructure (i.e. mechanical, electronics, general integration, shipping, processing, launch, etc.). Do $35 vs $350 make a big difference? How much does 300g of shielding cost? How much does an exta 300cc of shielding cost?

300g is an enormous mass compared to a standard Arduino.

A Nanoracks module has a mass of 1000g.  Addiing 300g to an Arduino module would make it closer to a Nanoracks module, might as well go with that standard platform, less development time, same mass.

Offline baldusi

  • Senior Member
  • *****
  • Posts: 8356
  • Buenos Aires, Argentina
  • Liked: 2539
  • Likes Given: 8273
Re: Arduinos and Raspberry PIs at ISS
« Reply #32 on: 03/03/2015 07:40 pm »
If I may give you my economist view, what the expected cost for the client, for the infrastructure (i.e. mechanical, electronics, general integration, shipping, processing, launch, etc.). Do $35 vs $350 make a big difference? How much does 300g of shielding cost? How much does an exta 300cc of shielding cost?

300g is an enormous mass compared to a standard Arduino.

A Nanoracks module has a mass of 1000g.  Addiing 300g to an Arduino module would make it closer to a Nanoracks module, might as well go with that standard platform, less development time, same mass.
Well, what's the target weight, power consumption, and overall cost?

Offline Danderman

  • Extreme Veteran
  • Senior Member
  • *****
  • Posts: 10288
  • Liked: 699
  • Likes Given: 723
Re: Arduinos and Raspberry PIs at ISS
« Reply #33 on: 03/04/2015 06:43 pm »
If I may give you my economist view, what the expected cost for the client, for the infrastructure (i.e. mechanical, electronics, general integration, shipping, processing, launch, etc.). Do $35 vs $350 make a big difference? How much does 300g of shielding cost? How much does an exta 300cc of shielding cost?

300g is an enormous mass compared to a standard Arduino.

A Nanoracks module has a mass of 1000g.  Addiing 300g to an Arduino module would make it closer to a Nanoracks module, might as well go with that standard platform, less development time, same mass.
Well, what's the target weight, power consumption, and overall cost?

Right now, the effort is to find the max number of such units that can be fit into a Nanoracks module. The question is also to optimize among 1U, 2U and 3U. In all cases, the max power consumption cannot exceed that of the equivalent Nanoracks module.

The business case does not close if the Arduino/Pi has a mass anywhere close to a single Nanoracks module. We need to pack 'em in.  This is similar to the concept when Nanoracks began, where a single Mid Deck locker would hold some number of Nanoracks modules - if the number of  Nanoracks modules that could fit into the MDL totaled 1, then we would have been dead.

« Last Edit: 03/04/2015 06:45 pm by Danderman »

Offline Prober

  • Senior Member
  • *****
  • Posts: 10348
  • Save the spin....I'm keeping you honest!
  • Nevada
  • Liked: 721
  • Likes Given: 729
Re: Arduinos and Raspberry PIs at ISS
« Reply #34 on: 03/04/2015 09:49 pm »
I am looking at the technical issues involving use of Arduinos as controllers at ISS. My opinion is that there may be a niche for Arduinos (and Raspberry Pi) class controllers for experiments, as they are physically smaller than Nanoracks modules, and we all know that it is expensive to transport hardware to ISS.

The broad objective is to develop a host for these units at ISS, subject to ISS system regulations, and bearing in mind that there is virtually no crew time available to play with these. They have to be Plug and Play and Forget.

My personal objective is to use an Arduino equipped with a Midi interface and a wireless radio to allow musicians on the ground to "play" in real time with anyone at ISS who has a Midi device (like a Midi keyboard), using direct line of sight from a ground terminal (when ISS passes over) and the wireless system. Since Arduinos support Midi and wireless radio, the ISS hardware could be pretty compact BUT Arduinos are susceptible to Radiation events, so I have to figure a way to reboot the Arduino whenever it crashes without the ISS crew spending all of their time sheparding the system.

Any help would be greatly appreciated - the Midi scheme might evolve into a KickStarter project for Midiots, and the generic host system might be viable as a business, you never know.

this kickstarter might have a controller you can adapt it uses midi
https://www.kickstarter.com/projects/staskevich/midiwidget-control-anything-via-midi?ref=category_featured

2017 - Everything Old is New Again.
"I fear all we have done is to awaken a sleeping giant..." --Isoroku Yamamoto

Offline ThereIWas3

  • Full Member
  • ****
  • Posts: 948
  • Liked: 500
  • Likes Given: 338
Re: Arduinos and Raspberry PIs at ISS
« Reply #35 on: 03/04/2015 10:18 pm »
One way to do it is to have the RPi's just do netboots, with no other software storage on board.  This would allow for quite small-capacity Flash drives with big rad-hard internals, lots of duplicated data, etc, just enough to get the netboot going.  On a power cycle, they send out a broadcast Ethernet message looking for a boot-host, which then downloads everything needed to run, parameterized by the hardware MAC address of the requestor.

This way only the boot server has to be particularly hardened, to support large numbers of simple RPis.  (Like hundreds of them).   Net-boot is a standard feature of most computer ROMs these days.  The boot server might also host persistent storage for any log files, collected data, etc.  And of course, the boot-server is itself duplicated.

And there are well-established programming tools for dealing with large networks of interchangeable processors that have the possibility of failing all the time.  (It just is not the usual C++ and Java stuff most people know)

Offline Danderman

  • Extreme Veteran
  • Senior Member
  • *****
  • Posts: 10288
  • Liked: 699
  • Likes Given: 723
Re: Arduinos and Raspberry PIs at ISS
« Reply #36 on: 03/05/2015 02:50 am »

Offline Danderman

  • Extreme Veteran
  • Senior Member
  • *****
  • Posts: 10288
  • Liked: 699
  • Likes Given: 723
Re: Arduinos and Raspberry PIs at ISS
« Reply #37 on: 03/05/2015 02:53 am »
One way to do it is to have the RPi's just do netboots, with no other software storage on board.

Yep, this is certainly the right track.

If this thing ever got to the point where a dedicated fixture were possible, one place to put it might be a in crew quarters equipped with a water wall. That might be enough to mitigate against upset events.

Offline Robotbeat

  • Senior Member
  • *****
  • Posts: 39270
  • Minnesota
  • Liked: 25240
  • Likes Given: 12115
Re: Arduinos and Raspberry PIs at ISS
« Reply #38 on: 03/05/2015 03:17 am »
And the nice thing about netboot is that if the cards burn out, you just replace them from a stock of identical cards. ...but they're very unlikely to burn out if they have only, say, 10MB of storage which is still 1-2 orders of magnitude greater than an Arduino but 2-3 orders of magnitude less than a full Raspberry Pi system.

The only issue is that the Pies use much more power than an Arduino and are also much, much larger than the very small SMT-based Arduinos, so as long as what you're doing can be done on Arduino, you're still better off with that.

Well, that, and you STILL need to harden the file server.
Chris  Whoever loves correction loves knowledge, but he who hates reproof is stupid.

To the maximum extent practicable, the Federal Government shall plan missions to accommodate the space transportation services capabilities of United States commercial providers. US law http://goo.gl/YZYNt0

Offline Danderman

  • Extreme Veteran
  • Senior Member
  • *****
  • Posts: 10288
  • Liked: 699
  • Likes Given: 723
Re: Arduinos and Raspberry PIs at ISS
« Reply #39 on: 03/05/2015 06:08 pm »
There is going to have to be an optimization process to determine how many Pis vs Arduinos can be hosted.

Another path is to look at the Intel Galileo board to see if it could perform more robustly in orbit.

« Last Edit: 03/05/2015 06:14 pm by Danderman »

Tags:
 

Advertisement NovaTech
Advertisement Northrop Grumman
Advertisement
Advertisement Margaritaville Beach Resort South Padre Island
Advertisement Brady Kenniston
Advertisement NextSpaceflight
Advertisement Nathan Barker Photography
1