Activity Feed › Discussion Forums › Strictly Surveying › I finally had one
I finally had one
Posted by Bruce Small on November 11, 2014 at 5:52 pmAfter years of surveying and probably several million shots, I absolutely had a bad RTK shot yesterday.
I took a shot of the beginning of the fence and walked to the end of the fence, under a frail palo verde tree about 30 feet south. It was one of those flimsy trees I wouldn’t worry about because an tenth or two was no big deal. I remember the shot well, and I didn’t code the shot until I got there so I couldn’t have pushed the button early. It plots about 4′ east and 5′ south of the first shot. Clearly wacko. I looked at the raw data in the collector and that’s the way it was recorded.
It happens, but the odds are in our favor, apparently.
mike-marks replied 9 years, 6 months ago 12 Members · 13 Replies- 13 Replies
Kent is dancing with joy
He can hardly wait to send you an “I TOLD YOU SO” letter.
It happens once in a while even with other types of equipment. The key is that your professional experience and checking sense caught it.
It really helps that I was the one on the site who took the shot, and I’m the one who is doing the drafting, so I’m the one who saw the funny line and heard the alarm bells go off.
I’ve had a few over the years. Most of the time its when I walked near a tree or building. I’ve also had it happen when doing topo on a 4 wheeler, you drive under a tree or something else and get a bad fix. Its easy to keep driving and getting bad data. I think the newer equipment is better than the stuff I had 10 years ago.
> It really helps that I was the one on the site who took the shot, and I’m the one who is doing the drafting, so I’m the one who saw the funny line and heard the alarm bells go off.
Many of us don’t have that advantage. And it truly is an advantage.This is the maybe not the first one. It is the first one you have become aware of. As I posted last week, it is also possible to get bad EDM shots as well.
I can pretty much count mine on one hand too! Jp
> … huzzah! 😉
:gammon:
> After years of surveying and probably several million shots, I absolutely had a bad RTK shot yesterday.
>
> I took a shot of the beginning of the fence and walked to the end of the fence, under a frail palo verde tree about 30 feet south. It was one of those flimsy trees I wouldn’t worry about because an tenth or two was no big deal. I remember the shot well, and I didn’t code the shot until I got there so I couldn’t have pushed the button early. It plots about 4′ east and 5′ south of the first shot.Okay, the problem is clearly with your checking practices. I have read that RTK never gives a bad shot, so the checking is completely unnecessary. If you cut back on the checking, I’m sure that you will find that the bad shot problem goes away as well. :>
Very interesting.
I’ve been using RTK since 2002 starting with the backpack Trimbles, Javads, Sokkia GSR2700IS, Trimble R8, Sokkia GSR2700ISX and now Topcon Hiper V. The only major brand I haven’t used is Leica. From 2007 onwards they have been GPS & GLONASS units exclusively.
I would easily get a dodgy fix once a week, sometimes multiple times in one day.
Admittedly I push the boundaries of the unit in vegetation but with longer, repeated observations with repeated initialisations (caused by my hand on top of the antenna rather than the software button to do the same) I do not have issues. If I can get four observations that agree within 20mm with repeated initialisations and then check them at a later date and get the same answers I know I’m OK. I don’t check every point at a later date but apart from some that I knew I was REALLY pushing my luck, I haven’t found any that I have had issues with.
My immediate first thought was that you haven’t been doing your checks if you have only just realised you had a dodgy initialisation. My second thought was that you must work in very open terrain with no (or very little) obstructions. I sincerely hope it’s the latter.
:good:
You are correct lots of open sky for me and if I had to push it under trees or next to buildings or other structures I pay close attention to the initialization and do all the tricks like losing initialization and staking out the collected points or initialize on a known point in Trimble for a check. I have been using rtk since 98. What I forget is the number of people who are not correctly trained or don’t have the discipline to be a surveyor. Such as the crews that get initialization in open sky then run under the tree before they lose lock to get the fire hydrant shot. Or the crew I saw the other day using rtk with a 5′ rod and both of them were over 6′.:-( When I take a measurement I can stand behind it, doesn’t matter if I did it with a right angle mirror and 100′ rag tape or a one second station. Depends on the job and tolerances required as to which equipment to use under what circumstances. What amazes me is the number of surveyors that are scared to try new tools because of stories they hear about bad initializations. In the meantime other disciplines and trades are embracing the technology and constructing roads without survey stakes. Go figure! My 2 cents! Jp
I’ve had more than few over the years, nearly always where there is some kind of a vertical surface nearby that reflected a clean enough signal for long enough that it was incorporated into the fix solution. I’m lucky in that I do all the field work as well as drafting so these things tend to jump up and bite me. Were someone else collecting the data for me I’d likely just assume it was what it was. After 15 years of using RTK I’ve developed a 6th sense for these gooobers, although it can be totally unpredictable. If in doubt, loose lock and stake it out. If it involves control, post process the PPK vectors. RTK is Murphy’s favorite hunting ground and around here, it’s a target rich environment. 😉
Willy> After years of surveying and probably several million shots, I absolutely had a bad RTK shot yesterday.
>
> It plots about 4′ east and 5′ south of the first shot. Clearly w[h]acko.Well, if that’s the worst shot you’ve had in 2 million shots, you’re into 6 sigma territory concerning accuracy overall, +- 5 feet.
Better is to consider it a blunder. The only blunders that damage a survey are those which are not detected and discarded. The problem with satellite surveying is sometimes blunders are not detected without repeat measurements, or offset stations if the site is poorly visible, etc, when multipath, foliage, EMI, etc. are present. Close in (a few thousand feet) a simple total station is much more accurate than a GNSS receiver trying to acquire weak signals and resolve under heavy cover with a lot of multipath.
Log in to reply.