Facet & Facet L-Band - PPK

GPS/GNSS Enclosed Products
1

Hello,

Is it possible to do PPK with the facet and the facet L-band. The L-band isn’t getting enough accuracy where we are and we don’t have cell service for NTRIP so we were thinking of PPK’ing some stuff. We don’t need the RTK as we are only collecting 4-8 points per site we visit for a 3D Laser scan. All the tutorials are for RTK. I think I have to turn on SW Maps logging for this to work correctly. Is it really that easy? Can I post process the data through Emlid Studio?

Thanks,

Eric

2

Oh, should I use the L-Band as the base? Will it help get the correct base position faster?

3

Hi Eric,

Obviously there are many ways to accomplish this, but I like to work the problem backwards.

What will the deliverable (I assume the 3D Point Cloud, or a derivative) require in terms of precision and accuracy ?

Examples:

  1. If the point cloud is mainly for visualization purposes for site or equipment planning, then using the L-Band as a Base and the Facet as the Rover would be my choice as the quick way to situate the 3D model in real world coordinates (possibly just an L-Band under certain situations).

  2. However, if you plan to use the 4-8 points as permanent control points (with targets) for scan-to-scan registration of the model as constraints, then I would take a different route. We’d typically use a total station for that.

The complexity is normally driven by what the data will be used for.

If you are able to share the requirements, then I’d expect you would get some detailed feedback and suggestions to consider.

4

Hi Ryan,

We have the L-band version. We tried that and we are too far North. There is no RTK service available where we are at. We would like to situate the model with some accuracy (2-3cm would be ideal). We do not want to use a control point from a total station. We are going the PPP route because of the no L-band, and RTK service. I have figured out how to get the PPP data from csrs-ppp, and there maybe a monument in the area’s we are working with as well. I tried using the “Mark” feature on the rover and it created a .CSV file but I have yet to be able to align this with the Rover raw data processed by Emlid.

5

Haven’t figured out how to use emlid studio but was just successful using RTKLib b34. Quite a few gotcha’s along the way:

  1. CSRS-PPP when using RTKConv needs to have a RunBy/Obsrv/Agency field filled out or CSRS-PPP won’t process the data.
  2. Don’t use the interval checkbox when processing the data in RTKConv or RTKPost.

From this we were able to to get results that matched our 3D Laser Scanner on a large 100m property in the range of 10mm-30mm.

1 Like
6

That’s a good result, considering what you are comparing.
I guess you really don’t know what portion of that 10mm-30mm is from scan registration, GNSS errors, etc.

Also Scale Factor now comes to mind. When comparing GNSS to terrestrial marks- a Grid to ground conversion of the GNSS coordinates will give you a better representation on how well the two are in agreement. However, the Scale factor might not matter much over a 100m site and it looks like you are in pretty good shape already. It’s just something else to consider if you wanted to drill down a little bit further for 3D model validation :smile: