Our company has two RS2 units and we’ve tried twice now to use the Stop and Go methodology unsuccessfully to get FIX quality PPK point locations.
I’ll describe the second attempt, as that has fewer variables:
Basestation
One RS2 setup on a tripod over a known point. We had the tripod in the same place for 5 days and have derived a location for that setup from AUSPOS (all 5 days produced positions within 1cm of each other), so even though it was unknown at the time, it is now definitely well known!
Logging was set to 1Hz and in UBX format
Rover
The survey site was approximately 16km east
The rover was set to log at 1Hz and in UBX format
The record time for the points in survey mode was 5seconds
The target was trees, so I was holding the rover up against the trunk of a tree for 5 seconds (no tripod/bipod was used)
We use the Stop and Go functionality in Emlid Studio to attempt to get FIX positions for each of the tree locations, but didn’t manage to get them all as FIX despite the rover showing +/-50cm accuracy from >30 satellites for the duration of the survey.
Is there anything we’re missing? Any field techniques we should be doing to improve the chance of this working in the future?
When doing Stop and Go, you need to collect very good raw data on the field to be able to have the FIX status in PPK. I’m afraid that, if you were against the trunk of a tree, then data quality of the rover is not there, because trees tend to attenuate the signal a lot. Where you above the trees or under canopy ?
The accuracy of 50cm you observed in real-time is not a guarantee that you will be able to obtain a FIX during post-process, because the computation process is totally independent.
When doing Stop-n-go survey, I would suggest first to have a 5-min stationnary data collection at the beginning and at the end of the survey, in an open-sky location. Then, when moving with the rover, be careful when you drop a significant number of satellites or when the PDOP gets close to 3. If that happens, another quick stationnary session should be done again.
Florian gave some good advice, so be sure to check I’d add one short note: we recommend setting higher Update rates for moving rovers. 5 Hz is usually enough.
I wouldn’t recommend using coordinates collected with a Single solution. To calculate the precision value, you can multiply the delusion of precision by three RMS. If the RMS is 50 cm, the precision value will be higher.
As for the base unit, it doesn’t move, so the 1 Hz update rate should work well.
I’d add one short note: we recommend setting higher Update rates for moving rovers. 5 Hz is usually enough.