RTK Part Zzzzzz
> Yes, refusal to test is a perfect out.
Well, considering that my equipment has been extensively tested, I assume that you must have one of the RTK users in mind. So far, the main answers to uncertainties of RTK-derived positions have been:
- add RTK vectors with covariances to LSA in combination with other measurements to verify weights and do error propagation (sophisticated answer)
- check coordinates from RTK solution by repeat occupation (less sophisticated answer that doesn't characterize uncertainty well)
- check RTK solutions from time to time and post those that agree well (no characterization of uncertainty)
- insist that it is RTK for crissake, so it is good! (Alfred E. Neuman methodology)
- ask why anyone would care what the uncertainties in RTK-derives positions are (widely used gambit held over from compass and chain days)
RTK Part Zzzzzz
> > Well, I don't think "trust" figures into land surveying. If you do, you may have missed your true calling in real estate sales. Just saying.
>
> You are exactly right Kent; just because [stuff]
This isn't some complicated point. Professional surveying requires maintaining some level of process quality and getting results that have known quantified uncertainties.
Anecdotal examples of the sort contained in the original post don't get you there, except at the beer joint. So, what is the alternative? It's having reliable estimates of the uncertainties in the positions to begin with.
That is what would have enabled the poster to know what the uncertainties in the azimuth and distance that he inversed between them were. I take it for granted that this isn't Rocket Science since I've been able to do it on every survey I've made for more than twenty-five years. It's really pretty amazing that I'm even having to point this out.
It isn't a subject that should threaten any professional surveyor in the way that is consistently proven to be the case whenever the subject of evaluating the uncertainties of RTK-derived positions pops up. That tells me that there are a whole bunch of RTK users who are simply operating on trust or blind faith.
Kent, you make me mad
Because there is truth to your complaint. And, I experience it every day.
I have my own QC mechanisms.
I am an RTK user.
I observe stuff, and re-initilize.
I re-observe.
I take an extra shot, 3-10 feet away, and compass and tape between them.
I feel that I have a common error ellipse, of around 0.05'. Sometimes around 0.12'.
In the mountains of Arkansas, this is Un-Godly accurate. When the total station traverse sometimes closed by 0.31' around 100 acres or so.
Kent, it does bug me that I have to take TIME to quantify my errors. But, that really does not bug me. What bugs me, is I sometimes find it lying to me. 4' bust. 3' bust. Even 7'. I figure it lies to me about 1 time per 100 shots. More in deep woods. Less in the clear. Horrible up near trees.
My point is that there are users taking ONE shot, and leaving the "Scene of an accident".
I have to agree. Some folks are abusing RTK. Me sometimes too. But, I do feel that over all it is a good solution. I rarely double check my shots, in the middle of a 20' wide creek.
But, the evolutionists, believe in MAGIC. "It did it just right, in impossible situations". RTK does it more often than Evolution!
Have a great day, and this has been a good thread.
Nate
Kent, you make me mad
> Kent, it does bug me that I have to take TIME to quantify my errors. But, that really does not bug me. What bugs me, is I sometimes find it lying to me. 4' bust. 3' bust. Even 7'. I figure it lies to me about 1 time per 100 shots. More in deep woods. Less in the clear. Horrible up near trees.
There are two separate issues, though. One is just trapping gross blunders and the other is characterizing uncertainties from normal random errors. The gross blunders do seem to be a feature of RTK use in "high productivity" (meaning: get'er done!) mode when trees are involved. Those 3 ft. busts that I found in an RTK survey from last year at two key corners of a tract were pretty clearly the result of some surveyor being pressed for time and using RTK in a setting where a prudent surveyor would not.
I think that is the Saturday Night Special aspect of RTK. Many folks are a lot more dangerous with it in their hands.
Kents Tests RTK
regularly.
Every time he retraces a "Dippity RTK Abuser".
And, he uses TRIMBULL equipment.
N
Kent, you make me mad
Kent, Keep it up.
FIXED at 0.06' means just that. It's within 0.06'. Probably more like 0.02'!
(Wink)
N
RTK Part Zzzzzz
> > Well, considering that my equipment has been extensively tested
>
> When did you get an RTK base-and rover setup or tie into a base with RTK to do such testing? Forgive me if I am reading this worng but you have always made it sound like you do not, and have not used RTK.
News flash! The reason I'm quite sure about errors in RTK surveys I follow is that I check them with higher accuracy methods, meaning a combination of post-processed GPS vectors and conventional survey measurements.
The higher accuracy of those methods is why I mentioned them as a means of testing the positional accuracy of RTK on an array of real world points in clear and semi-obstructed locations.
If you think RTK users typically use single-base RTK over distances much greater than 1 mile, then the separation between the base and the points in the test array should be increased to whatever distance you think is normal.
Kent, you make me mad
I completely agree with the need to properly communicate the results of our work. Part of that is the use of standard notation and valid statistical tests. Failure on this point is a major contributor to our lack of prestige as a Profession. Another major factor is refusing to engage in productive dialogue...
Over the last 15 plus years I have used Trimble, Ashtech, Leica and Topcon RTK systems driven by half dozen or so field software packages. Not one combination provides an out of the box QC number or default error estimate that reflects reality. That doesn't make them evil, it just means I test it and develop real world estimates. I follow-up by determining what the relationship between the quality indicators and true uncertainties are. There isn't much different in that process than the testing of any other equipment I have used over the last 35 years.
I have developed procedures that reflect the errors I know to be present in my RTK equipment, and catch the blunders I know occur. Mine isn't the only method that works but I dang sure know it does.
While I recognize there is a large segment of our Profession that does not know how to do this with RTK, I contend that ignorance is not found in any tool it is simply revealed through its improper use. For that matter no professional inadequacy is the fault of any inanimate object. If a surveyor will use RTK in ignorance they will do so with a 25 foot tape.
Few that come here do it only to toot their own horn or limit discussion through dogma and degrading comments. Don't let yourself be known that way. You are too smart and hold too much value for that...
My .02, Tom
Kent, you make me mad
Mr Bionic Tom,
If that was directed at me, then I comment back: I was kidding Kent. Kent is a very stickler for an error estimate. One that is real, with real numbers to back it up.
He is not experienced with RTK, except with the practitioners, who are abusing it. Kent is quite useful, with his cocky opinions. Because for me, he's preaching to the choir. But, I run into bad work sometimes too, by sloppy practitioners.
I love static GPS, because it yields the kind of redundancy that I love.
The combination of Static RTK is coming my way, in about 2 wks, via Shawn Billings... And I am looking foreward to it!
🙂
N
Jim Frame
I think you will like using PPK once you try it, I think you are using Trimble so it is fairly simple to use.
Some things to know is that you need to set-up a survey style to make it work correctly and be sure your base data and rover data are in sync. In other words the base needs to be collecting data at the same time as the rover, it doesn't have to be the same rate however, your base can be collecting at 1 second and the rover at 5 and you are just fine, but if the base is collecting at 3 seconds or 15 seconds and your rover at 5 you will have problems.
Also you will want to set fix times for your survey, it defaults to 8 minutes which means you need 8 minutes of collection time before you can fix your survey, doesn't mean you can't collect data before the 8 minutes, but if you lose lock you will have to start over, I dial down that time to 3 minutes for 5 or more satellites.
I will wait the 3 minutes before starting, usually, however sometimes I will start a location before it fixes, you just need to be very careful. Once you collect data t download it and it will act basically like a static session, after all that's what it really is, you can tie them to a Cors point, your base, whatever data is available. You will be quite surprised at the quality of a 3 minute PPK location. Do one and tie it to a base, then download a local Cors station and try it, it will open your eyes to some possibilities. Use different location times and try different Cors points, test it for yourself. Also the continuous topo function will return excellent results. Even to a far flung CORS point. For that RTK is safer, but PPK works well, you just don't know if it did until you get back to the office.
The problems using it are that you need time for it to fix, you need to stay locked as long as possible, you don't know if you are getting good data until later.
Definition of Saturday Night Special - Kent's Wrong
> > Night Special was a cheap weapon of inferior quality.
>
> No, a Saturday Night Special was something you bought on the spur of the moment to be dangerous. Same deal with RTK in the hands of the folks in a hurry.
The phrase Saturday night special (SNS) is a slang term used in the United States and Canada for any inexpensive handgun. Saturday night specials have been defined as compact, inexpensive, small-caliber handguns with perceived low quality;[1] - From
Wikipedia
BTW - my 10th edition of Merriam-Webster's defines it thusly: "a cheap easily concealed handgun"
Definition of Saturday Night Special - Kent's Wrong
Now you've done it. Prepare to be corrected :>
Definition of Saturday Night Special - Kent's Wrong
> > > Night Special was a cheap weapon of inferior quality.
> >
> > No, a Saturday Night Special was something you bought on the spur of the moment to be dangerous. Same deal with RTK in the hands of the folks in a hurry.
>
>
> The phrase Saturday night special (SNS) is a slang term used in the United States and Canada for any inexpensive handgun. Saturday night specials have been defined as compact, inexpensive, small-caliber handguns with perceived low quality;[1] - From
> Wikipedia
>
> BTW - my 10th edition of Merriam-Webster's defines it thusly: "a cheap easily concealed handgun"
So, in what way is your definition of a Saturday Night Special not equivalent to "something you bought on the spur of the momeent to be dangerous?"
