Thanks!
Thanks!
Thanks!
You're right! A couple of hundredths horizontally isn't going to make a difference. I'm more concerned about vertical control. In any case, I once read "Let the purpose of the survey dictate your methods." Thanks for your 2 cents!
Thanks for sharing your thoughts!
Unfortunately we only use RTK. I always watch the H & V residuals. After 20 or 30 seconds they don't tighten up significantly. Your proposed method is interesting. What are your thoughts on using mission planning software? Thanks for replying!
Thanks for clarifying that!
We're using a VRS network. After 3 sessions on each point, I can see outliers on 2 points. Those particular points are not open 100% to the sky. I believe we had a PDOP of < 2 on all sessions. Thanks for replying!
I'm in the suburbs. No room for resection. Thanks for replying!
Thanks!
This is the VRS network we use. I'm trying to learn a bunch of GPS terminology. I have 2 choices for mount points:
1. Network RTK (MRS)
2. "SBL" Single Baseline
The GPS guru who maintains the L-Net told me either 1 or 2 will work for RTK, but don't switch back and forth between them once you start your sessions. Does someone's choice of mount points depend on how far they are from a base station?
?ÿ
I found an excellent post concerning how to calculate a weighted mean at https://surveyorconnect.com/community/surveying-geomatics/weighted-mean/. Posted by RADAR on September 23, 2016.
Generally speaking, yes.
Single baseline solutions will use only one of the control stations to model atmospheric error and calculate corrections to broadcast to your rover. The RTK vectors stored will flow from that station. But the further you get from a single base station, the less likely the atmospheric conditions are to be similar between the rover and base. So you need to be close to that station to use SBL.
MRS (Multi Reference Station) uses several (minimum of three or five I think) stations to model the atmospheric error at your rover's location, generate a "Virtual Reference Station" right at your rover, and all your RTK vectors will flow from that virtual station. This is the main reason why you don't want to switch during a session, to keep your vectors from the same station (or virtual station) and using the same correction source.
If you are physically located "inside" the network (i.e. not operating on the edges), the MRS solution is usually the way to go, since you are surrounded by stations and the corrections are a composite of multiple stations observing all around your area. If you are close in to a single station and on the edge of the network, single base might work better.
A proper weighted mean has nothing to do with thoughts, feelings or preference. Weights should be assigned based on the demonstrated objective value of attributes and computed in a supportable manner.
Weights can be applied subjectively, but that is far from the only way to do it.
Hence the qualifier 'proper'...
Thanks for the excellent explanation! I took the list of L-Net reference stations and indicated the distances from Orlando.
According to some documentation I found, there are 4 reference stations that could fall in the recommended area of operation for a single base station.
I would expect nothing less from thebionicman.
Either way, it is the best professional opinion based upon observed and recorded factual data.
What is the purpose of this exercise? It gets to a point to where you're looking for accuracies that will not hold up to the standards of practice when you look at the requirements for differential leveling.