I have recently collected the location of a number of points using RTK with base and rover GNSS recivers (Emlid RS2). My base station was not placed over a known point, meaning the base postion was averaged and thus to my knowledge means any of the points collected are accurate to within several meters. I have downloaded the data logs for both the base and rover, and was wondering whether it is possible to improve the locational accuracy of these points using some form of post-processing technique. I have read about PPP and PPK but as far as I am aware these are used as an alternative to RTK as opposed to an improvement. I am new to all of this.
1 Answer
If I understand you correctly, the position of your base unit was determined by averaging uncorrected Base station positions, then the Base unit supplied RTK corrections to the Rover, is that right?
In that case, the spatial relationship between your collected points would be good (with RS2 hopefully around centimeter-level if you were working in good GNSS conditions) but you will have some fixed offset between the set of Rover features and the actual positions which is the same as the offset between the averaged Base position and the actual Base position.
If you have the ability to post-process the Base station files (I don't know anything about post-processing RS2 files) then this could give a more accurate base position. Simplistically, the offset between your averaged Base position and the post-processed base position can be applied to all your Rover positions as a Latitude / Longitude / Altitude shift.
Alternatively, you could go back to wherever your Base station was set up and determine that location more accurately (maybe using your RS2 Rover and some external RTK corrections, available on the Internet free in many countries / states). Again, the offset between your original averaged Base position and the new RTK-based base position can be applied to all your Rover positions as a Latitude / Longitude / Altitude shift.