I chose to cancel rocket testing this weekend and work on my GPS project.
I got the Rakon GMM8652 based front end working.
So now that I have a fully working front end the real work starts.
I'm currently routing the raw data into an FPGA , but instead of doing the processing there I'm just using that as a path to record the data. Recording raw data records (4M bytes/sec) and post processing with the FASTGPS open source software GPS program on a PC.
Next Step do the corelator processing in the FPGA. I have that verilog code written, but its mostly untested at this point.
Rough order of events:
- Prove the RF and FPGA hardware works. (done)
- Get one correlation channel tracking one sat.
- Get 6 or more channels tracking sats.
- Calculate position, velocity, time (PVT) based on this tracking data.
- Add data delay buffer and feedback from IMU data to aid loop tracking.
- Fly the thing.
and fly that so we have a good raw data set to use for testing and simulation.
14 comments:
Surprised there's an open source navigation engine & an open source RTK system, but no-one did the obvious.
That is interesting. Can the resultant code here be released as open source so all the pieces are available?
I am assuming your DIY GPS and IMU will not be hindered by MTCR restrictions. Correct?
~T
Paul,
I have been reading through your old blog posts and see that you've switched GPS hardware multiple times since you started Unreasonable Rocket. I'm still working through your old posts, but so far haven't seen an explanation as to why it was/is beneficial/necessary to invest all that time, money, effort, etc into replacing a component that is/was (apparently) already working.
I'm curious, what was missing, not working well, etc on your previous GPS units?
Thanks,
Thom
When we started flying the helicopter most GPS showed flaws.
5Hz units were found to do a lot of averaging, and if your using it for a control loop averaging is bad.
So we tried a bunch of units then upgraded.
We then switched from a L1 only to a 5K RTK L2 for accuracy for the contest.
After the contest qwe have been looking for a low cost light weight high dynamic unit. We don't need mm accuracy for the next step, we do need good IMU gps integration with very small and light.
I'd also prefer not to pay 20K+ for an integrated IMU GPS unit.
(You can get then for less than that, but they are not the deeply integrated units)
Paul,
Thanks for the extra explanation on your GPS choices. I also found your post from 06 June 2010 which had a lot of detail on your GPS + IMU Master Plan - and answered my question about MTCR/COCOM restrictions.
~Thom
Is your project open/libre source and publically avaliable somwhere?
Btw i'm working on something similar that you with a SiGE chip, i still doing acquisition part but i think you can give some hints about what to look with fastgps to adapt the code for a custom sampling and IF Freq.
I'm using osgps atm, just seems i got few data to get some sats...
Can you please post the verilog code link for gps correlators implementation.
Thanks
can you give the verilog code of gps correlator.
Nice work on the rocket, im just starting my build on a RepRap Prusa. I will be keeping an eye on your blog !
Little doubt, the dude is completely just.
I was working in a GPS project in my company because we have some deliver cars and the boss needs a report about where is every car !
I think this is one of the most important information for me. And i am glad reading your article. But should remark on few general things, The website style is perfect, the articles is really excellent : D. Good job, cheers
Post a Comment