Hi everybody
I'm a new user, maybe I have stupid questions. When processing data from the DJI Phantom RTK drone, the point cloud has a difference of approximately 2 m x / y. The coordinate system used by the drone is WGS84, but which one? EPSG code? Thanks

Welcome Agisan ! The Moderator patiam is an expert in rtk model iSSUES , he ll sure help you

Assuming you did not use an external correction source aligned to a different datum, the geotags in your P4R data are Geographic (Lat/Lon) coordinates WGS84 datum, height above ellipsoid vertical (EPSG 4326).

Not sure what to make of your "the point cloud has a difference of approximately 2 m x / y" statement. Your point cloud is shifted relative to some know feature or benchmark?

patiam Posted at 12-17 11:24
Assuming you did not use an external correction source aligned to a different datum, the geotags in your P4R data are Geographic (Lat/Lon) coordinates WGS84 datum, height above ellipsoid vertical (EPSG 4326).

Not sure what to make of your "the point cloud has a difference of approximately 2 m x / y" statement. Your point cloud is shifted relative to some know feature or benchmark?

Thank you for the answer. I have to get to the coordinate system in Romania, Stereo70 (EPSG 3844). with the built-in translator of pix4d or dji terrae programs I still have approximately the same differences. The checkpoint is an official one from the local authorities. I also tried the option to transform the coordinates and transcribe the exif, but I still can't manage to the right position the pointcloud.

patiam Posted at 12-17 11:24
Assuming you did not use an external correction source aligned to a different datum, the geotags in your P4R data are Geographic (Lat/Lon) coordinates WGS84 datum, height above ellipsoid vertical (EPSG 4326).

Not sure what to make of your "the point cloud has a difference of approximately 2 m x / y" statement. Your point cloud is shifted relative to some know feature or benchmark?

My P4RTK is flying approximately 30M too high when I operate a Terrain Awareness flight.  I've been downloading the DEM from the USGS site and then clipping it using QGIS.  patiam, you have any pointers or workflow tips to get this thing to fly correct altitudes?
thanks

Has anyone compared the P4RTK position (e.g. on the ground accounting for lever arm offset) with an actual surveyed ground control point?  

mach1drones Posted at 2020-12-19 14:37
My P4RTK is flying approximately 30M too high when I operate a Terrain Awareness flight.  I've been downloading the DEM from the USGS site and then clipping it using QGIS.  patiam, you have any pointers or workflow tips to get this thing to fly correct altitudes?
thanks

Your terrain model needs to be ellipsoidal height. I am guessing yours is orthometric and your GEOID is about 30m.

xabierr Posted at 3-16 16:08
Has anyone compared the P4RTK position (e.g. on the ground accounting for lever arm offset) with an actual surveyed ground control point?

I've done a quite a bit of testing.  Lever arm should automatically be accounted for,  I suspect lever arm inaccuracies are very small, less than a few mm.  Greater errors come from focal length and correction source.

Wow, missed the traffic on this thread from some time ago.

(BTW, I'm not a mod, just another user).

@mach1drones, I think @Skyveyor112 covered your question.

@xabierr, yes, this has been done by many users and the position of the P4 RTK w/ RTK fixed while sitting on a benchmark is within a couple cm, as one would expect.

@AgiSam: if you're using the proper workflow in Pix4D it may be that  their conversion to your chosen coordinate system is not quite right. You might try converting to another more common one, and comparing the coordinates of your known features also converted but by some other means outside of Pix, or try exporting products from Pix4D in geographic coords and convert in a GIS instead of Pix (in other words take the coordinate conversion duties away from Pix4D)..