Having flown the Mavinci, Falcon 8, Trinity, eBee and a variety of other experiments, I find the DJI from our Chinese overlords to be far less ass clenching.

I've lost count how many times there's been random errors mid flight or the craft decides to just fly off. Trinity is like the three bears with the battery or ESCs either too cold or too warm, too much wind or some random transition error.

That's pretty much it. We use the Maptek Pointstudio software to "register" then colour by distance against existing LiDAR and GPS points as a check.

Agisoft Metashape is the most popular photogrammetry software in our neck of the woods.

End product is a Maptek Vulcan triangulation and georeferenced ortho which can be draped over the triangulation to make it look pretty.

Most mines are based on a datum that was current at the time of commencement so applying the conversion from WGS84 and ellipsoid height to the mines datum and geoid height usually comes in pretty close then the Maptek "registration" feature translates that extra bit to match the LiDAR.

It's mining, so data needs to be produced quickly but needs to be consistent with previous surveys. Some machines move 60k BCM over a 24 hour period so unless it's finalised and delivered in a couple of hours it starts losing its usefulness.

You can co-ordinate the mobile base by measuring a point on the ground and enter the reference coordinates, use an online CORS service, or if you have a GNSS receiver with the right features, you can connect to that and set up NTRIP.

Existing LiDAR is used to compare and adjust if required, must have sufficient overlap of good level data. If the incorrect settings are applied when reducing photogrammetry, it could cause surfaces to distort, (create a some effect or some other distortion). GPS points provide a check to ensure the surface accuracy is consistent across the project area.

If you're not understanding how and why these checks need to be done, I would suggest not going out and doing this as a way to get some easy coin. There's a reason we do a few years at uni for this job. If you provide data that contains inaccuracies or issues and you charged out for it you'll open yourself up a big can of worms.

The RTK on the m300 is pretty good. We don't bother setting out GCPs for mining due to the frequency of different surveys and ever changing environment (anything survey gets destroyed by mine progression, especially if it's a peg).

We use existing LIDAR surfaces to compare and adjust if required. Driven GPS points on accessible roads then graded by surface offset is another check to ensure no distortion in the surface.