PPK processing for LiDAR and/or photogrammetry generally requires the user to provide the coordinates of the phase center of the antenna, rather than the marker location over which the antenna is set up over, at least with DJI Terra and ROCK Desktop I have found this to be the case. It's an annoying manual step that is required to get the best results using strictly PPK, but I always do it anyways by applying the height measurement plus the phase center offset for the antenna. Sure you can use GCP's to adjust your processed data later, but I find that 9/10 times PPK is very good. I frequently have LiDAR data checking to less than 1cm vertically compared to check points, using just PPK processing this way. To help reduce this manual step I made a web application to take the RINEX header information and spit out Latitude, Longitude and height at phase center, see https://sgss.ca/tools/rinex-base.html If anyone else is experiencing this pain point, hopefully this is useful.
NOTE: This was created with the help of an AI coding tool, tested against my previous more manual methodology using tools I programmed myself.
NOTE 2: The RINEX file header will need to have the final coordinates of the base point already in it, e.g., exported from TBC or Infinity, etc.
I'm really shocked that DJI's software does not incorporate antenna calibration modeling. A proper model allows for much more than just phase center differences. It also accounts for the potential signal distortions based on the signals horizontal and vertical angle of attack. It might be a bit overkill, but everything helps when you get one epoch to get it correct.
In the last 30+ years I have not seen a GNSS processing software that did not incorporate antenna modeling. The list of post processing suites I've had the pleasure, and some cases mis pleasure, to use includes include Trimnet, GPSurvey, Trimble Mapping Office, Trimble Geomatics Office, Trimble Total control, Leica Geomatics Office, Leica Infinity, Novatel's Inertial Navigator and NGS's Pages. They have all had some form of modeling.
That being said, give the software limitations your app is great help.
@john-putnam I don't know whether or not the DJI or ROCK Robotic software use the antenna modeling for the actual processing or not, I suspect they would (how could they not), but when you provide your base station RINEX file for the PPK processing, they do not account for the height measurement or the antenna phase center offset to determine the base location to process the trajectory, they read just the XYZ ECEF coordinates and assign them to be at the phase center unless you manually override the coordinate values. The other information is right there in the header, but it is ignored.
@leegreen Thanks for the report. I only tested a small number of antenna varieties, maybe it's the longer name that's tripping it up in this particular instance, will investigate and get that fixed. I'll send you a DM to request a sample file from you.
Regarding the UTM / State Plane coordinates, this was a last minute add-on to show UTM coordinates - as a quick check since grid coordinates typically are more meaningful to surveyors, but of course UTM is not universally used. I'll see about auto-identifying state plane zone but want to stay away from user-selectable grid projections, that is beyond the scope of the intended use case for this tool and adds extra complexity for the user when the intended output is really the latitude, longitude and phase center height. Another tool like my Coordinate Calculator | Simple Geospatial Solutions might benefit from an option to select a RINEX file though ... I'm planning on modernizing that tool at some point as well.