"I don't have it in me to argue with your irrational opinions about RTK for surveying."
Really?? Then why would you even start this thread?
Less Filling? Tastes Great?
Shawn Billings, post: 383432, member: 6521 wrote: I began by observing individual epochs over a period of hours at several baseline lengths. I imported those epochs into a spread sheet and averaged epochs into groups creating groupings of 10 seconds, 20 seconds, 30 seconds, 60 seconds, 120 seconds, 180, 240, 300, 360, up to 15 minutes.
I then compared the radial residual of these clusters to my control coordinates (which were determined by various post processed techniques and compared to the average). I then calculated the standard deviation of these residuals.
So, what you are saying is that these tests on a particular baseline were done on just one day? At the face of it, that would be problematic right there, although you can at least estimate the variance of the control values by some means.
I compared the standard deviations to the sample set to see that they coincided, which they did.
There must be a different way of expressing the above. I think what you meant was that you checked your arithmetic by verifying that the pooled variance of the different subsets of n epochs was equal to the variance of the whole sample from which they were formed.
Then I graphed the accuracies over time loosely on graph paper to arrive at an inflection in the accuracy to time. Per my graph, after about 4 minutes the accuracy improved but only very marginally. Armed with this information, I then went back out and tested the actual 4 minute observations and found that my accuracy estimate held up at distances of 2.55 miles and 9.5 miles from the base.
The most important element in all of this, though, is how the control values were derived, i.e. whether they were derived from the same period of time during which the RTK epochs occurred, or not. If from the same period of time, one would expect that would tend to artificially deflate the variance.
The second important element is whether the variance of the N and E components were nearly identical or whether the scatter plot of the positions were markedly asymmetrical along some axis such as that parallel to the vector back to the base.
The improvement of 240 seconds over 30 seconds was about 15% reduction in standard deviation which may not be terribly significant to most, but as I mentioned earlier, there are other reasons to extend the observation time.
The next interesting test would be see how the most efficient obserservation time varies seasonally.
Jones, post: 383431, member: 10458 wrote: While I'm not for sure how other RTK systems work. But if you follow the Javad manuals as well as what Shawn and the rest of the five team recommend you are going to spend at least three minutes on every point were accuracy is important. I would suggest you look at some of the videos than Shawn and some of the others have made as well as log on to the Javad board and read some of the post. It may surprise you on how many folks are trying to use RTK the right way.
I guess I'm just nostalgic for the good old days when equipment manufacturers actually tested their equipment and didn't rely upon their salespeople for that. It makes the whole effort seem like the beta version is still on the market.
The samples were not on the same day. One control point (at 9.5 miles) was created using OPUS Projects, the other (at 2.55 miles) by static GPS from several years ago. I don't recall at the moment how the third was established. I also compared the overall average to the control group. Variations were quite small.
Kent McMillan, post: 383436, member: 3 wrote: I guess I'm just nostalgic for the good old days when equipment manufacturers actually tested their equipment and didn't rely upon their salespeople for that. It makes the whole effort seem like the beta version is still on the market.
I like the fact that the sales people are actually surveyors who use the equipment everyday. It's also nice to be able to throw out an idea on a forum and Mr. Javad himself says he will implement it in the next update. Maybe the fact is you are a dinosaur stuck 15 years in the past and don't really want to help push the profession into the future.
Shawn Billings, post: 383438, member: 6521 wrote: The samples were not on the same day. One control point (at 9.5 miles) was created using OPUS Projects, the other (at 2.55 miles) by static GPS from several years ago. I don't recall at the moment how the third was established. I also compared the overall average to the control group. Variations were quite small.
However, the positions of the control points had variances associated with them that should have been added to the apparent variance of the subsets of the sample. Assuming that the previously surveyed position was error free would artificially deflate the RMS error you calculated since the control coordinates were actually not error free values.
Jones, post: 383440, member: 10458 wrote: Maybe the fact is you are a dinosaur stuck 15 years in the past and don't really want to help push the profession into the future.
Why is it that so many RTK users want to congratulate themselves for using quickie survey technology while being unable to answer fundamental questions about the results they are obtaining? If this is what the future looks like, no thanks, I'm driving.
Kent McMillan, post: 383443, member: 3 wrote: Why is it that so many RTK users want to congratulate themselves for using quickie survey technology while being unable to answer fundamental questions about the results they are obtaining? If this is what the future looks like, no thanks, I'm driving.
I would go out on limb and state that many surveyors could not set down and do a least squares adjustment by hand. Or most could not explain in technical terms how an edm works. I will state that most RTK users know it's limitations. While we may use RTK because it is quicker the majority of us know when to leave it in the truck. You maybe right about the "quickie Dickie" surveyors abusing RTK, but there have been half ass surveyors since the beginning of time. Keep in mind the tool is never the problem it's the one behind it.
Jones, post: 383445, member: 10458 wrote: You maybe right about the "quickie Dickie" surveyors abusing RTK, but there have been half ass surveyors since the beginning of time. Keep in mind the tool is never the problem it's the one behind it.
Well, I think it's one of those things that if you build it, they will come. RTK was originally marketed in the parts of Texas familiar to me as a great way to survey quickly without having to know anything about much of anything. I think the tag line was "GPS Total Station".
Wow!
It was just 10+ years ago that we were RTKing locations of power poles, fire hydrants, repetitive and assorted topo shots, cross-sections and profiles of dirt, etcetera etc.
My my, how tech flies ahead unbounded that one can establish boundary corners now. Some with consternation saw this future that it would happen despite raising caution against manufacturer's claims.
I remember the Stumpwater gang and their efforts to test and push inexpensive LI ProMarks to the max. Never said much at the time but thought" Geezus Guys, just save your beer money and get L2 receivers".
My first GPS seminar was 27 years ago using Ashtech L2 box in a parking lot along the shores of Lake Hamilton in Hot Springs. AR.
So I guess now it can be proclaimed that Dr. J. Ashjaee has brought us to state of Nirvana for boundary surveying.
Nobody steal that Nirvana survey brand name because I'm hot on it tonight
Nirvana Boundary Consultants Dba NIrvana BC
(Still not too excited about standing out in the field during strong rain event mashing buttons on a box)
Kent McMillan, post: 383446, member: 3 wrote: Well, I think it's one of those things that if you build it, they will come. RTK was originally marketed in the parts of Texas familiar to me as a great way to survey quickly without having to know anything about much of anything. I think the tag line was "GPS Total Station".
My view is simple, RTK is an amazing tool to have in the tool bag. Are there abusers?; sure but a large majority of us are trying to use it in a professional way. Am I going to layout a house with it, of course not, but a boundary survey were it will save me five or six set ups with a negligible difference in accuracy I'm going with the LS.
from the comments, it sounds like there are lots of really awful surveyors in Texas.
MightyMoe, post: 383450, member: 700 wrote: from the comments, it sounds like there are lots of really awful surveyors in Texas.
Well, at least in and around Austin anyway!
:innocent:
MightyMoe, post: 383450, member: 700 wrote: from the comments, it sounds like there are lots of really awful surveyors in Texas.
I don't know about that. I read this message board and the sophisticated RTK users are a very slim minority who deal with survey results as vectors that can be adjusted in combination with other types of measurements rather than as just numbers.
It does make me feel good that there are so many good surveyors here. I'd say 95 percent at least.
Kent McMillan, post: 383442, member: 3 wrote: However, the positions of the control points had variances associated with them that should have been added to the apparent variance of the subsets of the sample. Assuming that the previously surveyed position was error free would artificially deflate the RMS error you calculated since the control coordinates were actually not error free values.
I don't think this is true. In my test I assumed all error was in the RTK positions and that my control was error free. If I had assigned error to the control as well then the total residual would be divided by the RTK and the control, reducing the error of the RTK.
Kent McMillan, post: 383419, member: 3 wrote: One common area where RTK falls apart is when it is used to place or locate boundary markers on small lots or similar situations where the distances between markers are well under a couple of hundred feet.
I agree, and wouldn't use RTK on a smaller lot survey. What I would (and do) use it for is on boundaries that require a section subdivision. I'll tie section corners with RTK and set two points at the project site within site distance as far away as possible (with a third to check into) and do the rest of the work with a total station.
I'll compare my subdivision with other surveys of record as a check.
Shawn Billings, post: 383465, member: 6521 wrote: I don't think this is true. In my test I assumed all error was in the RTK positions and that my control was error free. If I had assigned error to the control as well then the total residual would be divided by the RTK and the control, reducing the error of the RTK.
Actually, you are right that the variance of the control coordinates was present in the variances that you calculated by assuming that the control coordinates were error free. If you know the variance of the control coordinates, you can back it out of the apparent variance of the sample, but I believe that you'd want the variances reduced to N and E components to do that and then compute the total distance variance from the components after subracting (root difference of squares) the N and E components of the control coordinates. If the horizontal DRMS error of the control coordinates is 3mm, and the horizontal DRMS error of the sample was 8mm and the N and E components were equal, then the corrected DRMS error of the sample would be :
SQRT[8^2 - 3^2] = 7.4mm
MightyMoe, post: 383462, member: 700 wrote: It does make me feel good that there are so many good surveyors here. I'd say 95 percent at least.
Can you provide a link to this hidden message board where substantially all of the RTK users don't vapor lock when asked to compute the relative positional uncertainty between two points they positioned via RTK? I guess the others post here, huh?
Sometimes kent misses jokes.
🙂
