RTK and Relative Accuracy Part 1.
> Something isn't adding up. RTK = PPK. PPK with longer observation time = improved accuracy (fast static). RTK with longer observation time = no improvement in accuracy.
You'll have to read Conrad's post more carefully.
There are a lot of "used to's" that don't apply anymore. No excuse for lazy thinking. A surveyor should be knowledgeable enough to know when to use what tool to accomplish his work with the proper balance between efficiency and accuracy. Professional discretion requires, well, discretion. At what point is a total station better than RTK? At what point is static GPS more productive? These are worthwhile discussions. It's not just about shaving hundredths, but also about efficiency. If it takes half an hour to get the RTK to produce a few repeatable shots on a point in the sticks, how long would it have taken to find the nearest clearing, set a pair of points and traverse to the point with the total station? What are the trade-offs?
We recently did a topographic survey of about 40 acres with about 2,000' of road frontage (curb and gutter). The RTK made good sense in the open field. The robot made good sense on the curb and gutter. In both cases the accuracy and efficiency both pointed to using one tool in one scenario and the other tool in the other scenario. The robot can produce better verticals in much less time than RTK. For curb and gutter that makes sense. The RTK didn't require numerous setups that the robot would have to accomplish the ground topo. Also, the increase in vertical accuracy from the robot would have been insignificant on dirt shots.
Here's the thing. If not you, as a professional surveyor, then who? Who's responsibility is it to "strain out the gnats" of our profession? Who is supposed to understand the history of private land ownership and land dispensation, land law (particularly as it related to boundary reconstruction), advanced Euclidean geometry (2D and 3D), Earth sciences related to measurements on, above and below the surface of the Earth, the art of presenting all of this in a relatable way to lay-people? Shouldn't someone be having these conversations? Isn't that why we are the professionals?
RTK and Relative Accuracy & Static Accuracy as well
So while you guys are "discussing" the accuracy vs. precision of RTK/Static observations, we did a rapid static/static observation for over 30 open line traverse stations to adjust their closures. These open lines are not closed because closing them would take too long and had to pass by heavy traffic roads that I mentioned awhile back.
Anyway, we had 2 base & 3 rovers. The 3 rovers were dropped off one at a time and the 3rd one was observing for minumum of 15 minutes. So as the 3rd rover reached the 15 minute interval, the first two were probably recording data into to 20-30 interval.
So there are 2 data overlaps from the 2 bases, as well as 3 overlapping data from the 3 rovers.
When I processed the data using GNSS Solutions, I was separating it into 2 datasets - one for each of the base & compared their results. There was 3 points that were getting H & V variations well into the 0.40 m levels when I compared them with the traverse coordinates.
Could not determine why this was happening when other points were in the range of less than 0.10 m level.
Did all sort of data processing - 1 base vs all rovers; 2 bases vs all rovers; 1 base vs all rovers with no overlapping sequence between the 3 rovers;
Lo & behold, the source of the large discrepancy was due to the overlapping data between 3 rovers. Overlapping time between the 3 rovers were at maximum maybe 15 minutes. Sometimes less because even though all 3 rovers were on, it does not mean that they were recording data at their location.
When these overlapping vectors from the rovers were removed from the processing, the positions (H & V) came closer to the traverse dataset.
I was under the belief that the more overlapping data that you observed, the better your results. Apparently this is not true. I am not sure if the data from the overlapping rovers at less than 15 minutes is skewing the data from the bases? Vectors between rovers are shorter than vectors coming from the 2 bases.
That's the problem with these "black boxes", we sometimes (or most of the time) take their outputs as Gospel Truths because the mathematics behind all the data gathering & PP softwares are just way above our educational level.
Oh and did I mention the variations in output from the different PP softwares?
Static Relative Accuracy
Hello Kris,
> Pure poppycock. Static blows the total station out of the water at distances of 15 feet.
Well you sure got my attention.
What total station? Under what conditions? If you've got, or know of, some tests then I'd love to read them.
Static Relative Accuracy
> Hello Kris,
>
> > Pure poppycock. Static blows the total station out of the water at distances of 15 feet.
>
> Well you sure got my attention.
>
> What total station? Under what conditions? If you've got, or know of, some tests then I'd love to read them.
15 feet? I can probably do even better with my pocket tape, pencil, and paper. Reality check here