Can I somehow combine the benefits of on-site RTK and Post Procssing Static?
I have an upcoming job where I need to place some permanent markers.
I plan on using the following flow:
Set up the base on a roof with excellent sky coverage, overlooking the site I have to place the markers.
Connect to a NTRIP service from a station 1 km away.
Use the rover to collect static measurements from the ~10 different places. Baseline from my own station will be min 50 meters, max 200 meters.
And now for the tricky question: From experiments with RTKLib and using public data (and limiting the computation to use L1 only), I can get down to 2 mm 3D RMS with just 1 hour observations with the base station 200-300 km away (so quite a bit better than what I can get by just using RTK with NTRIP). I have to place 10 points, so I obviously can’t spend 1 hour on each marker. I can probably spend 5 minutes pr marker, tops, and 2-4 hours on the basestation.
The site is a bit tricky, warehouse buildings and foliage will challenge the sky coverage at some of the markers. Worst case I will only have direct visibility to a 90 deg x 80 deg view on one of the markers.
Using RTK seems to provide some resilience to losing the fix, but can this be obtained in the post-processing as well ?
If I use RTK and NTRIP as sketched out above, will it affect my raw collection opportunities and PPS-computing in a positive or a negative way?
Other side-effects, I have thought of ?
I want to have as much precision as possible, with the data collected on-site. The markers will be used for the next few years, so it will be worth spending some time on getting the precision right.
In your case, it seems more optimal to get the raw observation from Reach base and NTRIP reference station and to collect measurements with Reach rover using Survey tool.
Then, during post-processing, you can figure out Reach base coordinates using the reference station raw log. After that, you can post-process rover log against base log using these known base station coordinates.
If I only have part of the sky in clear view, with buildings close by (but with with base having an unobstructed view of the sky), then how long would you recommend to observe?
It’s difficult to answer the question correctly, cause the observation time is highly dependent on local environment and satellite visibility. It’s possible that 5-minute observation will be enough, but it’s also possible that 20 minutes will be insufficient to have good results.
There are two tools you can use for an indication in ReachView app.
The first is a signal-to-noise ratio chart that lists all visible satellites and corresponding SNR. When SNR of a satellite is over 45, it will be marked green. So it will be better to achieve as many satellites signals in “green zone” as possible.
Another one is a Survey tool you can use to store points. Survey tool provides points auto-saving, and there are some rules that can be configured to save the points fulfilling the requirements: DOP, precision and observation time. You can read more about it here.
Thank you for all your answers.
One last question, how does the SBAS sats affect the equation? Do they provide a position contribution, or do they only provide atmospheric corrections?
