Re: [gpsd-users] Support request for NavSpark GPS modules!

[Greg Troxel]

"Gary E. Miller" <address@hidden> writes:

> Looks like you need 2 units to get the cm accuracy?  One as a base
> station and one as a rover?  It says the base station sending RTCM 3.x
> data can be 10Km away, so I'm guessing maybe an internet base feed is OK?
>
> Would the project be to support both base and rover modes?

You may also be interested in:

  http://www.emlid.com/reach/

which is a tiny (really tiny; the photo is a person with small-mid
hands!) board with Intel Edison and Ublox M8T plus antenna, $600 for a
pair.  It runs rtklib and in theory does 5 cm relative positioning.
Either unit can be base or rover, and you can use public base stations
(e.g, my state DOT runs them).

A lot of this is built into rtklib, plus emlid has open-sourceb GUI
code:
  https://github.com/emlid/ReachView

Generally these units are not trying to export time to another computer;
they are just doing processing of carrier-phase data between units.

RTK these days seems to be about RTCM3, which I think is sending the
carrier-phase observables over a differential data connection.   I am
unclear on whether one needs the observables for a particular second to
calculate the rover position at that second, or if there is some
forward-reckoning like on regular DGPS.

signature.asc
Description: PGP signature