Staking out Surface

Emlid Flow & Emlid Flow 360 Getting started
1

In this topic: https://community.emlid.com/t/landxml-surface-file-sizes/39497 I asked about optimising Land.XML surface sizes. With the help from the community I’ve been able to do that well.

Subsequently I created a surface on the basis of 0.25m contours and imported it to Emlid Flow 360.

I then went back to the site and:

  1. Staked my GCPs and Checkpoints
  2. Traversed the surface and staked 60 points along a series of transects on the surface
  3. Took the results into QGIS as a set of points with elevations
  4. Sampled the DEM with those points to generate the elevations for those points in the DEM
  5. Calculated elevation deltas.

All points were FIX using an AusCORS station with a baseline of ~2.5km. RMS errors were around 0.012m to 0.013m in all dimensions.

The results of the GCP and Checkpoint stakeouts are:

Screenshot 2024-09-27 at 12.50.26
Screenshot 2024-09-27 at 12.50.26848×302 10.6 KB

The results for the sampled points on the surface were (negative numbers are a high stakeout, positive numbers a low stakeout - fill required)

Screenshot 2024-09-27 at 13.02.42

I guess with a 0.25m contour gap the maximum possible error approaches 25cm.

The surface file was only 3.5Mb or thereabouts, so I guess I could make much closer contours and see if the differences in elevation remain as large.

I’d be interested in views about these results.

What this exercise reinforced for me is the need to be able to produce a surface stakeout report in Emlid Flow just the same as you can for point stakeout. It’s a pain taking points back into QGIS to produce the same result.

2 Likes
2

And for completeness, here’s the result from 102 staked points on a surface created from 1cm contours and with the same parameters as the last set of points. Given this is drone data those contours are inside the GSD…

The differences between this and the 25cm contours seem to be negligible.

Screenshot 2024-09-27 at 15.39.42

4

Hi @MikeH,

I’m sorry for the radio silence. Thanks for the detailed explanation of your workflow and for sharing the results!

What this exercise reinforced for me is the need to be able to produce a surface stakeout report in Emlid Flow just the same as you can for point stakeout.

This is indeed a good feature to have when staking out surfaces. I’ll add this request to our internal list. Thank you for pointing that out!