dual antenna for heading

[Gregory Walker]

Has anyone used the Piksi RTK solution between two receivers fixed relative to each other?  I am looking at putting together a dual antenna "GPS Compass" and know what to expect from products that perform this function with a 1m or 2m baseline [distance between the two antennas].  Using Piksi and developing the necessary code to make the compass functionality is possible but if someone has already accomplished this task I would like to learn more.

Thanks.

[Clive Turvey]

Can't say I've done it with that short baselines, but certainly should be viable, or extensible to attitude. I'd personally favour a far lower cost solution.

https://groups.google.com/forum/#!searchin/swiftnav-discuss/clive$20cep/swiftnav-discuss/RYaT6XSjM4o/pMh5Bsv6L6AJ

If there's sufficient interest I might hook some Piksi's to the attitude array.

There has been some talk/suggestion that the baseline initialization should use a baseline length (easy to mechanically effect) vs baseline vector (NED) it requires now.

[Gregory Walker]

Could you recommend a lower-cost solution.  I know the solution needs to track the carrier phase of the two antennas.  From what I have found products that accomplish this are more expensive than a couple Piksi's and their antennas.  Plus, with the Piksi solution I hope to eventually use the RTK positioning for a rover - one of these two heading units would become a rover.

[Fergus Noble]

Hi Gregory,

I made an internal demo for an attitude system with 3 Piksis using a ~0.5m baseline. It worked reasonably well. The code isn’t suitable for release but it was really very simple to do. I just took the cross product of the two baseline vectors and normalized it to get a vector perpendicular to the plane the Piksis were in.

For heading its even easier, you can just do atan2 on the north and east components of the NED baseline.

Fergus

--
Fergus Noble

CTO & Co-founder

Swift Navigation Inc.
2148 3rd St.
San Francisco, CA 94107

--
You received this message because you are subscribed to the Google Groups "swiftnav-discuss" group.
To unsubscribe from this group and stop receiving emails from it, send an email to swiftnav-discu...@googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

[Michael Levy]

Hi Fergus,

What kind of radio modem did you use (RFD900+?). If it was the RFD900+ can you share how you wired it to the Pikisi?

I'm assuming the default firmware allowed you to use one base station with multiple (at least 2) Pikisi units - if not can you describe how you were able to do so.

Thanks!

Michael

[Clive Turvey]

The two Piksi's at the remote end only need one radio, Rover Master would provide observations directly to Rover Slave, allowing the Slave to determine its relative position, and the Rover Master would receive observations from the Base via radio to determine its absolute location.

I'm not sure the wiring is overly complicated.

UARTA(Master) TX goes to UARTA(Slave) RX and Radio RX (input to transmitter)

UARTA(Master) RX goes to Radio TX (output from receiver)

Grounds are common

Radio has it's own supply, capable of supplying required current

Sending Rover observation data to the Base isn't necessary

You could also use UARTB and configure it for the right SBP mask, if that's easier

[Rai Gohalwar]

Hi Michael,

We have a RFD900+ Integration Guide on our wiki. If you follow the setup shown, you should be able to use the nodal feature of the latest SIK 2.x firmware.

Just as a confirmation we have tested the node feature on Piksi. We were able to provide observations to multiple rover Piksis from one base Piksi using the RFD900+ modules. 

Best

Rai Gohalwar