CubeSats are nothing new – hundreds have been launched into Earth orbit by schools and universities over the past decade. Like anything cool, an Arduino eventually gets thrown into the mix. That’s what the folks behind ArduSat are doing: they’re launching an Arduino-laden satellite into orbit with a bunch of sensors to enable anyone to become a citizen space scientist.
On board the ArduSat is a suite of sensors including a spectrometer, Geiger counter, IR light sensor, electromagnetic wave sensor, a temperature sensor, gyroscope, accelerometer, magnetometer, GPS unit, CO2 sensor, and of course a few cameras. The rewards for this Kickstarter are fairly interesting: backers who pledge $500 are able to buy a week’s worth of time using the ArduSat sensors for your own personal experiment.
As for how this Arduino-powered satellite is getting a ride up to Low Earth Orbit, the team plans to send an application into NASA for the CubeSat Launch Initiative ride-along program. If NASA selects the ArduSat, it’ll get a ride into space along with other CubeSats on a larger commercial launch. If the ArduSat isn’t selected by NASA, the team behind this satellite has secured funding to piggyback on a commercial launch.
Tip ‘o the hat to [HackTheGibson] for sending this in.
Exactly how useful will cameras that are likely low resolution be in orbit? Things tend to be quite far away. I wonder if the entire space station would register as more than a pixel. The earth view could be good, though.
Hi ShortCipher! During normal flight, you’re right, the space station won’t be in view of the cameras, since we will be too far away and the resolution is too low.
The idea with the cameras is for people to take pictures at cool things they fly over – their home country, hurricanes, aurora, things like that.
Even in a low orbit I doubt it can work for more than 10 minutes (and I’m being optimistic). AVR is not radiation-hardened, nor it has internal error-detection mechanisms.
The idea is nice, but, please, note that up there you have absolutely no protection from high-energy stuff. And a single bit-flip can kill your project.
Alvie
As a communications system designer for a University Nanosat Team, AVR’s can indeed last in space. You just have to enclose everything in aluminum. It should last a surprisingly long time.
I’m one of the engineers on the project – user code will run in parallel on multiple processors, and then use voting to protect against random errors/bitflips.
They’ll also be cased in aluminum to help protect against ionizing radiation, and the power supply has built-in latchup/overcurrent protection.
How will they shield the AVR from ionising radiation?
There are no radiation hardened AVR’s so I assume they will somehow shield it?
Found answer to my own question:
Yes, for a while. One of the biggest challenges for electronics in space, besides all of the power and thermal issues, is radiation. The high-radiation environment can cause temporary short circuits (called single-event latchups), introduce errors in code (called bitflips), and eventually degrade electronics by altering the resistive properties of key junctions.
To prevent against bitflips, multiple Arduinos will run each program simultaneously, and use voting to eliminate random errors. To protect against latchups, the satellite will constantly monitor for over-currents and respond to protect the key circuits. And although ArduSat is too light to have any shielding to protect against long-term damage, the satellite’s relatively short lifetime (about 4-18months) means it will reenter before total cumulative damage to become a problem.
Pretty sure the damage to the code stored in flash of an unprotected AVR will quickly render it useless.
What if it flips an internal system bit? They are only protecting against program memory corruption.
Well that’s the problem right there. They don’t actually have space hardened equipment, they just have a bunch of CPUs which will spend a lot of time endlessly voting against each other. That is assuming they don’t all crash due to program memory corruption.
Hi, I’m one of the engineers on the project, sorry it’s taken me so long to get on here and try answer your questions!
In terms of space-hardened equipment, we’re only using AVRs for the payload processor bank – the main flight control computer, the one running all mission-critical critical systems on the bus, is actually a NanoMind 712C, which has flown successfully on a few CubeSat missions.
They have interresting solution:
“To prevent against bitflips, multiple Arduinos will run each program simultaneously, and use voting to eliminate random errors.”
:))
And that’s what they call a democracy, folks
I give it 5 minutes before the multiple arduinos suffer program memory corruption and lose the ability to vote against each other.
If only *people* who had “program memory corruption” would lose the ability to vote…
This is one of the worst applications of an arduino
inaround the world. They’re made for prototyping, not finished projects. Especially not for something so size, weight, and power critical as a satellite. All the extra headers, programming interface, multiple circuit boards, and code overhead is nothing but a drain.Hi Rachie, I’m one of the engineers on the project:
The AVRs on the satellite essentially WILL be used for prototyping – the reason we picked Arduino was because the goal of our mission is to build a platform on which anyone can invent, prototype, and test their own code.
The mission critical systems and “finished product” part of the satellite is not run on AVRs, but using space-proven components; the flight control computer, EPS, and transceiver all have flight heritage.
run our own code on it? Am I missing something or isn’t this all data logging, so given the data logged we could do whatever analysis we wanted?
I’m sorry but this is crap. There’s no reason to run any computations onboard this satellite. They’ll do much better simply streaming the data from all the on-board sensors back to earth for processing.
Furthermore none of their sensors are rated to work in space, and there’s zero chance they’ll actually make a 2013 launch if they don’t even know how many sensors there’s going to be on their non-existent ‘experiment’.
@Rachie I agree. Using actual Arduino boards for this is a bad choice. What bugs is the guys are engineers so it wouldn’t be that hard for them to layout new PCBs with everything on it if they wanted to. It’s a jury-rigged cob job held together by string and bubble gum in my book.
On the other hand, there is some novelty in sending an actual Arudino into space, so I think that the project is more about that than a serious experiment going into space.
Hi Frank,
Although the prototype we’re using on the ground is a basic setup with standard AVRs and breakout boards, we actually are printing the the flight model on a custom PCB, to bring down the weight, help with the thermal dissipation, and reduce the amount of connections that can wiggle loose/crack in the temperature gradients/etc.
As for the novelty aspect/why we picked Arduino for the payload, the goal of the project is really to make a platform for the general public to design their own code, and Arduino seemed the best choice since it had such a large community and user base, and has a relatively shallow learning curve for new users and students.
I find this entire project highly suspicious. My prediction is that it will not actually happen, and investors will get nothing for their money. Buyer beware.
I’m sorry you think so, are there any specific concerns/questions about the project I can try and answer for you?
The top photo seems to be the eastern Mediterranean showing Egypt (Cairo), Israel and Saudi Arabia.
This would be a lot more interesting if they were going to make a hardened arduino and send it and a regular one up in to space. Run the same program on both to also get data about how quickly the unhardened one degrades.
That’s a good idea – did you see the ideas contest we’re running?
Discover Magazine is helping us out with it, you can submit ideas until July 6, and they’ll pick the best idea to fly on the satellite and give the winner a development kit (working ground-version of the payload).
I don’t want to spam the thread, I’d prefer to just answer questions, but the contest rules explain it best so I’ll put the link here: tinyurl.com/DSCRulesv2
Wouldn’t it be easier if they put one large server/server rack up in orbit and virtual a bunch of arduinos à la amazon EC2?
For the pilot project, we’re sticking to using actual Arduinos as a proof-of-concept that it works, but emulating instead is an interesting route we’re thinking about for the next satellite – or maybe even fly both to demonstrate the performance difference between the two approaches.
I’m glad some of you were not at kittyhawk
I don’t get the clue of having accelerometers, vibration and shock sensors in the weightlessness and having a barometer and gas sensors in vacuum…
An update for those interested. We’ve been working with these guys and they now have two of their cube sats on the ISS waiting for deployment in late November 2013. They have a 3rd scheduled for launch on 8 Dec.