skycooler
Posts: 3
Joined: Sun Jan 19, 2014 7:58 pm

GPS module for high-altitude, high speed?

Sun Jan 19, 2014 8:07 pm

I am working on a project to send a rocket to space, and want to use a Pi for avionics. I am getting stuck with the GPS though. My target maximum altitude is 115 kilometers, and I cannot find a GPS module that is rated for more than 18 km. Also, a problem I have heard from several other amateur rocket builders is that their GPS stops tracking when the rocket is moving at high speed, so I am also hoping to find a GPS module that can track at at least one kilometer per second vertical velocity. Does anyone know where to look for such a module?

jamesh
Raspberry Pi Engineer & Forum Moderator
Raspberry Pi Engineer & Forum Moderator
Posts: 27422
Joined: Sat Jul 30, 2011 7:41 pm

Re: GPS module for high-altitude, high speed?

Sun Jan 19, 2014 8:57 pm

Try the arocket email list. There are people on there trying to do this.

Although the GPS is the least of your problems, 115KM is a hell of a long way up. I don't think it has ever been achieved with an amateur rocket. Even 100kft is a challenge for amateurs. Look up 'carmack prize', Armadillo aerospace, and STIG-B.
Principal Software Engineer at Raspberry Pi (Trading) Ltd.
Contrary to popular belief, humorous signatures are allowed.
I've been saying "Mucho" to my Spanish friend a lot more lately. It means a lot to him.

User avatar
0xFF
Posts: 202
Joined: Tue Nov 20, 2012 7:52 pm
Location: Poland

Re: GPS module for high-altitude, high speed?

Sun Jan 19, 2014 9:22 pm

AFAIK there is no a commercial GPS module you want.
It is a limit just to prevent exactly what you want to do - build your own rocket ;-)
Ask US army for GPS module without speed limit.

skycooler
Posts: 3
Joined: Sun Jan 19, 2014 7:58 pm

Re: GPS module for high-altitude, high speed?

Sun Jan 19, 2014 9:37 pm

jamesh wrote:Try the arocket email list. There are people on there trying to do this.

Although the GPS is the least of your problems, 115KM is a hell of a long way up. I don't think it has ever been achieved with an amateur rocket. Even 100kft is a challenge for amateurs. Look up 'carmack prize', Armadillo aerospace, and STIG-B.
Thank you for the advice! I know that it's a difficult challenge, and I have been working on getting it there. But whether or not I achieve it, I want to be able to see what happened. I looked up the Carmack prize, and in this page about the winners, it seems they also lost GPS tracking and thus had to use alternate methods to verify the altitude. So that confirms my point about tracking at high speeds and altitudes.

jamesh
Raspberry Pi Engineer & Forum Moderator
Raspberry Pi Engineer & Forum Moderator
Posts: 27422
Joined: Sat Jul 30, 2011 7:41 pm

Re: GPS module for high-altitude, high speed?

Mon Jan 20, 2014 9:02 am

skycooler wrote:
jamesh wrote:Try the arocket email list. There are people on there trying to do this.

Although the GPS is the least of your problems, 115KM is a hell of a long way up. I don't think it has ever been achieved with an amateur rocket. Even 100kft is a challenge for amateurs. Look up 'carmack prize', Armadillo aerospace, and STIG-B.
Thank you for the advice! I know that it's a difficult challenge, and I have been working on getting it there. But whether or not I achieve it, I want to be able to see what happened. I looked up the Carmack prize, and in this page about the winners, it seems they also lost GPS tracking and thus had to use alternate methods to verify the altitude. So that confirms my point about tracking at high speeds and altitudes.
It's a well known limitation of commercial GPS. The only use is for high speed high altitude is rockets, and generally manufacturers/governments are cagey about giving high accuracy data in case it is used for nefarious purposes.
Principal Software Engineer at Raspberry Pi (Trading) Ltd.
Contrary to popular belief, humorous signatures are allowed.
I've been saying "Mucho" to my Spanish friend a lot more lately. It means a lot to him.

ric_rpi
Posts: 206
Joined: Thu Jul 11, 2013 2:33 pm
Location: Bedford, UK

Re: GPS module for high-altitude, high speed?

Mon Jan 20, 2014 10:57 am

Surely your problem is actually 'What is my Altitude & Speed?' not why can't I use GPS.

You could try putting a transmitter on your rocket and measure the Doppler to get the velocity. But for your target altitudes you may run into issues with transmitter power & RF licensing.

If you really want to use a PI then you could attach some inertial sensors, however these will drift so the journey time would have to be short. (I don't know how long it takes a rocket to get to 100km)

Would something as simple as measuring the temperature (on the end of a probe?) be accurate enough for your altitude measuring requirements? https://www.fas.org/irp/imint/docs/rst/ ... 01_019.jpg

RH314
Posts: 8
Joined: Thu Jan 30, 2014 3:23 pm

Re: GPS module for high-altitude, high speed?

Thu Jan 30, 2014 4:07 pm

Commercial GPS receivers conform to COCOM limit, (altitude < 18km) and (speed < 515m/sec) cannot be simultaneously exceeded, either can be exceeded, for prevention of use in missile or ICBM. I’m not sure if there isn’t regulation prohibiting private individual from shooting rockets into space. If you are with an institution maybe can try to sign declaration with GPS chipset/module vendor ensuring it’s only for scientific study use, to see if they’ll lift the limitation for you.

skycooler
Posts: 3
Joined: Sun Jan 19, 2014 7:58 pm

Re: GPS module for high-altitude, high speed?

Fri Jan 31, 2014 3:33 am

ric_rpi wrote:Surely your problem is actually 'What is my Altitude & Speed?' not why can't I use GPS.

You could try putting a transmitter on your rocket and measure the Doppler to get the velocity. But for your target altitudes you may run into issues with transmitter power & RF licensing.

If you really want to use a PI then you could attach some inertial sensors, however these will drift so the journey time would have to be short. (I don't know how long it takes a rocket to get to 100km)

Would something as simple as measuring the temperature (on the end of a probe?) be accurate enough for your altitude measuring requirements? https://www.fas.org/irp/imint/docs/rst/ ... 01_019.jpg
Thank you for the ideas!

We are planning to use a radio transmitter as well; one of the members of our team is licensed as an amateur radio operator. But it's hard to get accurate results at such short (relatively speaking) distances, as an error of a single millisecond is an error of about 200 miles.

As for inertial sensors, I am experimenting with an accelerometer/gyroscope board, but it seems the drift will be significant during the ~10 minute flight.

The problem with measuring temperature is that the air is so thin that it has very little effect on a thermometer - that is, in the thermosphere it takes several minutes to get an accurate reading and we won't be there long enough.

I wonder whether it's possible to build a custom GPS receiver, without the COCOM limitations? I don't know enough about microchips to know whether this is feasible or not.

ric_rpi
Posts: 206
Joined: Thu Jul 11, 2013 2:33 pm
Location: Bedford, UK

Re: GPS module for high-altitude, high speed?

Fri Jan 31, 2014 9:12 pm

You would need specialized hardware to perform signal capture, correlation, tracking and generation of PRN codes and a few man-years of software effort to build your own GPS receiver.

However you may be interested in this project to find out more http://home.earthlink.net/~cwkelley/

User avatar
jbeale
Posts: 3716
Joined: Tue Nov 22, 2011 11:51 pm
Contact: Website

Re: GPS module for high-altitude, high speed?

Fri Jan 31, 2014 9:35 pm

If you really want to make your own GPS, it's "just" a software problem. Nowadays you can get all the RF hardware on a single chip, for example http://www.skyworksinc.com/uploads/docu ... 02445A.pdf ...but I don't think you really need to do that, if you just care about basic flight parameters and are not actually doing ICBM guidance.

The COCOM limit is high speed AND high-altitude (although I know some GPS vendors interpreted it conservatively and locked out on either high-speed OR high-altitude). As an example, the Adafruit Ultimate GPS has been shown to work at high altitude and low speed in balloon flights.

As far as I know, any good modern GPS design will lock on to satellites within a few seconds, when relatively stationary, from a hot start (meaning it has current satellite ephemeris, etc loaded.) So if the GPS was locked on before liftoff, I would think you could re-acquire lock quickly once the rocket reaches apogee and that tells you your position and peak altitude. Meanwhile, inertial sensing will give you a good measure of thrust while the engines are running, and the combination gives you a good picture of the whole ascent phase. On the way down, I presume you could design in enough drag to avoid a GPS cutoff at 515 m/sec (1152 mph = Mach 1.5 at sea level). I'm no rocket expert, though; I can't claim to have tried this myself.

There could be another choice, if you can find a GPS that works in raw-output-only mode. If it does not solve for position internally, presumably it cannot implement an altitude limit. You don't get the position output "live", but you simply run the recorded raw observations through the open-source toolset RTKLIB and determine the complete track without limitation, as a post-processing step. For best accuracy you should simultaneously record raw data from a known fixed base station, which should be within 10 km of your "rover" (rocket) so you can remove some errors due to fluctuating ionosphere and troposphere. But even without that, you can get the basic GPS accuracy of several meters.

If I understand it correctly, you will be able to get a pair of such "raw-output" GPS devices for $50 as part of a crowdfunding campaign which is running now through Feb.6 called "NavSpark" and you would want "NavSpark RAW". If you're interested, check out: http://www.indiegogo.com/projects/navsp ... /x/6094574 and also the NavSpark FAQ.

User avatar
jbeale
Posts: 3716
Joined: Tue Nov 22, 2011 11:51 pm
Contact: Website

Re: GPS module for high-altitude, high speed?

Sat Feb 01, 2014 3:23 pm

I asked Oliver from the NavSpark / SkyTraq project about your GPS application, and he did not sound encouraging.
Oliver Huang said 10 hours ago
SkyTraq tries to provide best in class performance for consumer grade GPS/GNSS. Applications exceeding COCOM limit are not intended applications. We don’t know nor have resource & means to ensure receiver performance under those extreme conditions. As each launch is costly and require considerable planning, it is best for the other GPS/GNSS chipset companies 20 times our size, who have resource and means to verify their receiver performance under such conditions, to support these rare uses.

ric_rpi
Posts: 206
Joined: Thu Jul 11, 2013 2:33 pm
Location: Bedford, UK

Re: GPS module for high-altitude, high speed?

Thu Feb 06, 2014 10:29 am

jbeale wrote:If you really want to make your own GPS, it's "just" a software problem. Nowadays you can get all the RF hardware on a single chip, for example http://www.skyworksinc.com/uploads/docu ... 02445A.pdf ...but I don't think you really need to do that, if you just care about basic flight parameters and are not actually doing ICBM guidance.
The IC is just the frontend and would need to be connected to a Digital Signal Processor to perform correlations and FFT functions to find the code and Doppler offset of a GPS signal. A high spec PC may be able to do this but the pi isn't fast enough so more hardware components would be required.

Is this a personal project or for a company?

User avatar
jbeale
Posts: 3716
Joined: Tue Nov 22, 2011 11:51 pm
Contact: Website

Re: GPS module for high-altitude, high speed?

Thu Feb 06, 2014 5:37 pm

Actually I suspect the Pi hardware could do it- it's just that you would need access to the GPU :-). And a lot of knowledge about GPS and signal processing. I'm pretty confident a modern desktop GPU would have plenty of power for the problem, if it can be done by DSP hardware in a handheld GPS unit running on 3.3V at 30 mA so I think the hardest part is "just software" and that would be assuming a fix is needed in real time. Presuming that you are not trying to do real-time guidance of your rocket, you can simply record the received data for processing at your leisure afterwards. Just thinking... the GPS L1 C/A code is modulated at 1.025 Mbit/sec, let's say you record signals from your RF frontend at 16 Mbit/sec to capture 8 MHz bandwidth. That's within the bitrate of compressed HD video, and we know the Pi can record that. (Maybe 2x that if you capture I/Q separately, but it can still be done even without using both I and Q). Here is a brief overview of what you can do with such a recording: http://www.acasper.org/2011/11/07/gps-signal-analysis/

By the way, there is an updated FAQ about the NavSpark which has some relevant info: http://www.skytraq.com.tw/Commonly%20As ... stions.pdf
The COCOM limit is: (18km altitude) and (1000knot or 515m/sec speed) must not be both exceeded simultaneously or the GPS will not give valid result; it’ll still work correctly if either is exceeded. Thus if speed is Mach 1 (340m/sec) and goes up to 80km altitude, as it’s still within COCOM limit there will not be software imposed limit making GPS not work.
[...]
Modern day GPS chipset/module performance are not too much different, they all can withstand 4G acceleration and provide continuous fix. I don’t suspect there is fundamental issue with our GPS handicapping it for hobby rocket use. The possibility of application issue is higher.
Last edited by jbeale on Thu Feb 06, 2014 6:06 pm, edited 1 time in total.

User avatar
mikronauts
Posts: 2817
Joined: Sat Jan 05, 2013 7:28 pm
Contact: Website

Re: GPS module for high-altitude, high speed?

Thu Feb 06, 2014 6:05 pm

Would a pressure sensor not suffice for a reasonably accurate altitude?

Or triangulate from 2-3 RF ground stations?
ric_rpi wrote:Surely your problem is actually 'What is my Altitude & Speed?' not why can't I use GPS.

You could try putting a transmitter on your rocket and measure the Doppler to get the velocity. But for your target altitudes you may run into issues with transmitter power & RF licensing.

If you really want to use a PI then you could attach some inertial sensors, however these will drift so the journey time would have to be short. (I don't know how long it takes a rocket to get to 100km)

Would something as simple as measuring the temperature (on the end of a probe?) be accurate enough for your altitude measuring requirements? https://www.fas.org/irp/imint/docs/rst/ ... 01_019.jpg
http://Mikronauts.com - home of EZasPi, RoboPi, Pi Rtc Dio and Pi Jumper @Mikronauts on Twitter
Advanced Robotics, I/O expansion and prototyping boards for the Raspberry Pi

Return to “HATs and other add-ons”