Duane Frymire, post: 383259, member: 110 wrote: I am aware of all that. What I have trouble with is the idea or a method of applying it to rtk. In a static network it would be usual to have all stations similar in that observation times are alike and physical environments are all picked for good sky view with little or no multipath. Therefore you can get a realistic random error. Multipath being the greatest random error, and differing considerably from one rtk position to another, I'm not sure that least squares including all those in a network is a realistic representation. I question the "standard" way being applied to rtk positions, they simply don't have the commonality required for the analysis.
That is exactly what you'd get out of adding conventional measurements between points located via RTK and adjusting them in combination. That is, the uncertainties of the conventional angle and distance measurements should be well characterized in any situation where they would be made, so large residuals in the RTK positions as adjusted identify unrealistic weights. For the purposes of rigorous treatment, you could divide the RTK vectors into variance groups, to separate those generated under unquestionably good conditions from those that aren't. If it turns out that if the same scalar can be applied to the processor estimates for both groups, then you'd have a basis for concluding that the additional effort to separate them isn't necessary and all you're really interested in is trapping blunders.
BTW, Trimble specifies the RTK accuracy of their R8 receiver as:
Horizontal Accuracy +/-1cm + 1ppm RMS
Vertical Accuracy +/-2cm + 1ppm RMS
Are there really any surveyors who can't measure between two points with an uncertainty less than +/-1cm RMS? I would have thought that it would be obvious that there are all sorts of other ways to get a much more accurate answer. Ditto the vertical accuracy of RTK.
MightyMoe, post: 383265, member: 700 wrote: I tell my clients all the time that I'm 5% uncertain, makes them feel so good.
Who wouldn't want that 5% discount you're giving them?
Kent McMillan, post: 383254, member: 3 wrote: LOL! If you want to claim that you can make a survey more accurately via RTK than via any other surveying technology, be my guest. That would be more fun to watch if you weren't a Texas licensee, though.
I know it's difficult to admit that you are espousing knowledge, as an authoritarian no less, about equipment you've neither used nor tested, but believe me, it is helping your personal growth. Keep at it tiger! You're making us so proud! Look how far you've come!
Kris Morgan, post: 383278, member: 29 wrote: I know it's difficult to admit that you are espousing knowledge, as an authoritarian no less, about equipment you've neither used nor tested, but believe me, it is helping your personal growth. Keep at it tiger! You're making us so proud! Look how far you've come!
Hey, that's the manufacturer's specification. Your argument is obviously with Trimble. If you want to set grades on structures with RTK, I'm sure we'll all be glad to hold your beer and watch.
From a personal growth standpoint, it's disingenuous, and hampers the growth process, to have a myopic view of DIN specs which are at a single observation. Obviously, multiple observations and adjustments to quality points can tighten up most networks. The same can be said of the journey you're on to personal growth. The more experiences we have (observations), relayed against previous experiences (quality points), then we can begin to grow by evaluating them against one another (adjustment in networks). Hopefully this metaphorical lament helps.
Still though, one must first admit they've never experienced something to grow, and then gain that vital experience to maintain growth.
Kris Morgan, post: 383282, member: 29 wrote: From a personal growth standpoint, it's disingenuous, and hampers the growth process, to have a myopic view of DIN specs which are at a single observation.
I'm afraid that your argument is still with the Trimble people if you object to them specifying the accuracy of their RTK gear by baseline accuracy.
RTK accuracy per Trimble for their R8:
Horizontal Accuracy +/-1cm + 1ppm RMS
Vertical Accuracy +/-2cm + 1ppm RMS
If you say that you can spend a week getting different RTK solutions to two points 100 ft apart and actually come close to approximating the accuracy that one measurement with a good quality total station can easily deliver, you've basically admitted that RTK surveying is an inherently low accuracy method that requires lots of remeasurements to even approximate what can be done with higher accuracy survey methods without all the screwing around.
Not at all. It's all about using the right tool for the job. 100' apart is faster and more efficient with the total station. It is, however, not as fast or efficient when I'm measuring across the tract that would have taken a dozen traverse points. So, as you have often said, those DIN specs are at a 1 sigma level, or roughly 68%. As we all know, two sigma or roughly 95% is widely considered to be the "industry standard" for precision and accuracy. That being said, multiple observations allow for those data sets to be tightened up. Three, one minute, observations would yield 180 observations, under different initialization would be more than sufficient for any data set as promulgated by an statistics text book. Add those multiple observations, all without having to post process those values, in conjunction with an network adjustment, and I will gladly say that my RTK points are very close to what a static observation of 10 minutes would be (that would be the recommended time for ANY short baseline under 10km or 6.2 miles, regardless of it's length under that amount), and while you're still waiting, I'm moving forward with speed and efficiency. I know what points need more than one observation. I also employ a delphi method of analysis (widely used within a segment that believes it to be a part of the scientific method), for determining these things.
So, in closing, after beating you over the head with the obvious, your personal growth will not continue, if you do not step outside of your comfort zone. Even the NGS employs and uses RTK, and they set the standard (literally) for how and what to use. Of course, if you cannot see past your own limitations, then your mileage may vary. These posts used to make me angry with you. As I get older, I realize more and more, that you're part of a subset of humans that, despite their considerable intelligence, cannot evolve, due to in my opinion, a fear that they will not have mastery over the subject immediately. This would explain outdated and archaic equipment, methods, software, and even that old early 80's carry all you rolled. No, at this point, I simply feel sorry for you because at your advanced age, you will not change but simply deteriorate into a much more insolent and sad person. I also feel sorry for those whose company you keep having to deal with a passive-aggressive Type B, but I would wager that group is smaller and smaller each year.
You cannot live in fear.
Lee D, post: 383288, member: 7971 wrote:
Yes, the spec that I quoted above is for a GPS R8 and the GNSS-enabled R8 quotes slightly better accuracies for static and RTK, roughly in proportion to those quoted for the GPS-only unit. Note that for both the RTK accuracies are markedly lower than for static and fast static, as one would naturally expect.
The above specifications prove two things:
1) As usual when it comes to RTK, Kent is wrong. The spec is 8mm Hz and 15mm V
2) As any experienced user knows, static is much more precise than RTK
No one in their right mind would use RTK to measure something that had to be as accurate as possible at 100'. RTK is another tool in the tool box. The above specification is plenty good enough for almost anything. Do you really believe that a 600' topo shot in one face with a total station is better than an RTK topo shot? Do you shoot every topo shot with multiple sets in each face, tie them to more than one control, and then perform an adjustment?
If people aren't comfortable doing boundary with RTK, I can appreciate and respect that. Personally, I'm not comfortable with it for a lot of things either. But I know what I'm getting with it, I can have confidence in it, and I can defend it if necessary.
A smart-azzed attorney can pull that nonsense about having personal knowledge of the software, algorithms, etc., but in this day and age the use of technology and the reliance on manufacturer's specifications is widely accepted. In fact, the use of the GulfNet VRS for elevation determination in Louisiana is mandated by statute.
When I look at an RTK solution in TBC, I'm looking at error estimates at the 95% confidence level. These are based on the variance / covariance data from the measurement engine. If I have static or conventional observations on those points, I can combine them in an adjustment using weighting groups.
If you take two (or more) RTK observations at different times, those observations are completely independent of one another, with the exception of the base if it is still on the same setup. But the base is easily blunder checked before any other measurements are made by simply using a fixed height tripod and then checking into another control point. When a point is measured more than once, the software can display both a variance and a standard deviation in northing, easting, and elevation. If two (or more ) measurements are made at different times from different satellites with completely different geometry, it is completely reasonable to believe that they represent the correct values, within the standard deviation of the measurements. I guess if I had to, I could go find a PhD in mathematics to point out that the statistical likelihood of two independent fixed integer solutions producing the same wrong answer is, for all intents and purposes, zero. But then, if you understand what a fixed integer solution is, you already know that.
Based on a lot of these posts, we must be the only surveyors in the world who do anything other than boundary and construction that requires sub-millimeter precision at all times.
Lee D, post: 383292, member: 7971 wrote: The above specifications prove two things:
1) As usual when it comes to RTK, Kent is wrong. The spec is 8mm Hz and 15mm V
2) As any experienced user knows, static is much more precise than RTKNo one in their right mind would use RTK to measure something that had to be as accurate as possible at 100'. RTK is another tool in the tool box. The above specification is plenty good enough for almost anything. Do you really believe that a 600' topo shot in one face with a total station is better than an RTK topo shot? Do you shoot every topo shot with multiple sets in each face, tie them to more than one control, and then perform an adjustment?.
Actually, both Trimble brochures you and I linked show that I was correct in pointing out that RTK is inherently less accurate that many other surveying techniquies, including Static and Fast Static that one can do with the same receiver. It's obviously true.
What the RTK folks consistently fail to admit is that the main reason for using RTK is that it is fast. That explains perfectly why so few RTK users can give anything resembling a professional account of the uncertainties inherent in their surveys made with RTK. That's not to say that there aren't some sophisticated RTK users who actually do a professional job of adjusting different categories of measurements, including RTK vectors, but the evidence of this thread would be that they are a slim minority.
Lee Green is on the right track.
Here are the specs on the Javad:
http://javad.com/downloads/javadgnss/sheets/TRIUMPH-LS_Datasheet.pdf
It is a 4 page data sheet. In pdf format.
Go to page 2 for the specs.
For those of you who are not familiar, right click the link, and tell it to "Open in another tab".
thank you
Nate
Nate The Surveyor, post: 383302, member: 291 wrote:
Here are the specs on the Javad ...
You'll note, though, that the Javad specification doesn't assign a confidence level to the uncertainties quoted, which is at best an amateurish error and at worst an intentional misrepresentation.
The earlier versions of the RTK specs for the same receiver are quoted at +/-1cm Horizontal and +/-1.5cm Vertical (unknown confidence), so one suspects that the "new, improved" accuracy spec may not be full disclosure.
Kent McMillan, post: 383253, member: 3 wrote: for purposes of land surveying the 95% confidence levels are certainty.
I know that 95% is a very commonly used value, but I can see this dialog happening:
Surveyor: I use 95% confidence limits to check that I have accurate results.
Lawyer: So what is the chance that a measurement is off more than you recorded.
Surveyor: 5%
Lawyer: So if you measure 20 lines in a subdivision, you'd expect one of measurement to be wrong and it might be my client's boundary that you measured wrong?
Surveyor: Well, ...
Lawyer: No further questions.
Back at post #53
James Vianna, post: 382965, member: 120 wrote: RTK ... its a "calculated/insured risk". Whether you decide to assume this risk based on your experience or beliefs is up to you.
I am envisioning the following exchange:
The exchange with the lawyer might read no differently if you substitute EDM for RTK. They're both measurements in my book, they both have probable errors that can be quantified, and in neither case is a calibrated baseline on the client's site to check against.
Bill93, post: 383335, member: 87 wrote: I know that 95% is a very commonly used value, but I can see this dialog happening:
Surveyor: I use 95% confidence limits to check that I have accurate results.
Lawyer: So what is the chance that a measurement is off more than you recorded.
Surveyor: 5%
Lawyer: So if you measure 20 lines in a subdivision, you'd expect one of measurement to be wrong and it might be my client's boundary that you measured wrong? No further questions.
No, it goes more like this:
Surveyor: "The measurements that I show on this map meet accepted standards for professional surveying and I'm satisfied that my work as presented on this map contains no significant error and is for practical purposes correct and reliable."
Lawyer: "Okay, you were present at the first trial, were you not?"
Surveyor: "Yes was present for part of the trial and in fact did testify upon certain matters."
Lawyer (in best Perry Mason "got'cha" voice): "Do you recall having seen this [flourishing what apparently is an exhibit from the first trial showing measurements by another survey that corroborate what Surveyor has just testified]?
Surveyor: [carefully examining paper] "No, Mr. Lawyer, I don't believe I've ever seen this before."
Lawyer (trying not to appear deflated): "No further questions."
Kent McMillan, post: 383294, member: 3 wrote: the main reason for using RTK is that it is fast.
I believe this is true, and that static will produce a more accurate result (though often not by much) most of the time. But "fast" in this case can mean more than just observation time on a point; it can also mean getting results in real time that dramatically speed up the process of finding the points one wants to position. The latter -- being able to find boundary marks in real time, rather than having to post-process control point data and then return to the site to search for monuments -- is the most valuable aspect of RTK to me.
Similarly, RTK is very useful when locating monuments in heavy traffic areas. Even with multiple site visits to obtain redundant observations, this can be a lot faster than setting static control, post-processing, then returning to locate the monument with a total station, particularly if multiple monuments spread out over a large area are involved.
I still use my RTK gear sparingly, as most of my work requires greater accuracy than RTK can practically provide. But for some projects, it's a very effective tool.
