GPS Compass

[apagg]

Has anyone tried to make a compass using two GPS receivers like the NEO-6m or similar and putting them say 1m apart?  It looks to me that this would be possible if the accuracy was good enough

Regards
Andreas

[baltika_no_9]

I'm wondering why anyone would even try. How do you envisage this acting as a compass?

[apagg]

I'm wondering why anyone would even try. How do you envisage this acting as a compass?

I was thinking this was the way GPS Compasses like https://www.simrad-yachting.com/en-gb/si...s-compass/ worked. I see now that they don't use the position, but some kind of phase difference between the GPS receivers.

[baltika_no_9]

Right I see. But things like that are around £1000 when you could buy a 10DOF sensor for less than £10. You'd also have to write your own software libraries once you'd understood the theory behind the functioning of such things..

However if you really need very rapid updates and marginal improvements in accuracy then fine. I'd be interested to know how you get on. But I would be surprised if they are (m)any individuals who have undertaken such a project and come up with a reliable working solution. Don't think I'm knocking your idea, it's very interesting and I'd love to understand how it's done but quite a piece of work I suspect.
Good luck!

[steved]

Hello,

Wow! That sure brings back memories. Yes, spent a few years attempting to develop that a few years ago. Perhaps technology has changed and now it is viable. It was called a vector system and exactly like you described. Two antennas apart to be able to give a direction. (Why it was called Vector) In the end, it was a bit of a disaster. Sounds simple in theory....but the devil is in the details.

It does work however to work well corrections need to be broadcast and implemented in real time to both of the receivers. Not really something attainable on a boat that continues to travel. (Moreso getting accurate corrections)

We ended up dropping that system and instead developed an inertially coupled GPS system capable of RTK-F positions in real time with spotty coverage. (multipath and terrain)

[mgrouch]

To achieve accuracy it would require 2 multi band GPS antennas (not usual ones)
with support of up to L5 GPS bands and multiple GNSS systems (GPS, Galileo, GLONAS, etc)

GPS receiver needs to be able to send raw gps data (again not average GPS receiver chip) to RTK Library

RTK library seems does support moving base and a rover combination of receivers,
but I do not think it’s Kalman method implementation would take advantage of fusion of data from
different bands

Overall looks complicated and not too cheap for hardware with no guaranteed success.

More links on the topic you can find in this discussion on BBN forums

https://github.com/bareboat-necessities/...ssions/340