With recent GPS advancements and the continual questions of GPS working well under canopies, what is the general opinion of shots in trees?
One of our crews brought this back recently, within the last 6 months or so. In the particular shot, the unit was "fixed" and took approx 200 epochs to store. It is a Leica GS14.
In the photo, the GS14 is fixed with coordinate qualities, "3DCQ" of about 0.10
Given the photo above only, would you accept the shot?
Since the photo, the job has been resolved and returned to for other work.
If you are talking about a single initialization (no matter how epochs), then NO!
Loyal
This is a link I came across via another survey page, https://www.linkedin.com/pulse/texas-surveyors-you-ruining-your-profession-allowing-field-horsburgh referencing the degradation of survey quality due to poor procedures..
Loyal, post: 451902, member: 228 wrote: If you are talking about a single initialization (no matter how epochs), then NO!
Loyal
Personally, I was upset when I was sent to this particular job and found that these were the conditions of the control.
It's hard to explain to some folks that "fixed" doesn't mean "good or right or accurate."
When it take several minutes (or 5 or 10 minutes) for the receiver to "fix," under less than ideal (or good) conditions, that is your first clue that things are IFFY.
Even under IDEAL conditions, a second fix (dump the head) is mandatory. In IFFY conditions, then 3 or 4 fixes that agree within a centimeter (0.03 FT), is a reasonably good indication that the fix(es) are "accurate." This is especially true when you are talking about fixes that span 10-20 minutes. The likelihood of getting identical (+/- 0.010 meters) solutions with that kind of time span (that both stink), is pretty slim.
Loyal
In my shop we have a lot of GS14s. For ties to property corners, we often will pick up a couple RTK shots (broken init between) and also a short static session. Thank goodness for redundancy, because we've gotten some scary results. We've had situations where both RTK shots were fixed and agreed with each other, but differed by feet from the fixed post-processed result. So that's two independent RTK fixes and a static fix. In some cases the RTK was right and the static was wrong. In other cases, NEITHER ONE was correct. Like 99% of the survey community, we use RTK day in and day out, but (with apologies to a certain Austinite), RTK ain't all it's cracked up to be...and sometimes neither is a fixed static baseline. And while these are a very small minority of shots, that pucker feeling sticks with you.
We've also had some phantom height issues (with either the GS14 or GS15, can't remember which now), where a few tenths would suddenly show up for no good reason. We checked positively everything, and we really did lay the blame at the receiver's feet. The first time I've felt that way after 17 years of using GPS from Trimble, Ashtech/Thales/Magellan, Sokkia, and earlier Leica. We've been catching these errors because we require redundancy and systematic check shots, etc., but it really makes me want to have a drink of that Javad KoolAid.
Brian McEachern, post: 451901, member: 9299 wrote: Given the photo above only, would you accept the shot?
Not without ample verification. If RTK were the only option, I'd likely want to get at least a 3-minute shot on each of 3 different days and different times of day and have them all agree within tolerance (which would depend on the project) before I'd call it good. But since RTK is rarely the only option, I'd probably find another way of getting the measurement given the crappy sky view in this circumstance.
I would not accept this shot. It looks like form the photo that there are more 'clear' areas where I can set two control points with the gps, walk back to the truck, grab the robot, and locate the point and expect a more reliable result. Of course I'd still check those two control points as described above.
I have older GPS equipment (HIPER GGD). So I simply do not trust the GPS in most situations, since I live in the northeast where tree canopy is almost a daily issue, I don't use the GPS as much as those who live in other parts of the country.
It's a time suck waiting fifteen minutes for some sort of solution and swatting mosquitoes.
Not enough information. Loyal is correct and Jim Frame for that matter. You need multiple fixes spread out over time. Also what are the accuracy requirements? Is this the only point to collect in the area? Is it a small site or large area? I take shots like that almost every day. But I might not if the job requirements could not be efficiently met with rtk.
The Trimble R10 doesn't show a "fix" any more. It just shows a horizontal and vertical precision. If the precision doesn't get below what you've setup it won't store the point. Sometimes 0.1ft. is good enough but more than is setup in the data collector to be acceptable and then you can force it to store the point. Multiple shots usually just verify the same precision. I don't know why you would be setting control in a spot under a tree. That's not even a good place for an instrument setup.
We started everything back when we were doing a lot of static work with multiple crews setup in different spots and long occupations.
Considering that we have some good choices for RTK we rarely set up and do a static observation.
That being said, we locate each point with RTK for 60 seconds, dump the pole and collect another observation. Later that day we return and get two more observations.
We rarely see differences since we are typically in wide open areas but this is a great (and easy) check. Again it is not often but every once in a while we will see a deviation between the first observation and the one seconds later, or from the afternoon batch.
We do this with Topcon receivers and Leica although while I like the Leica equipment, I think Topcon has a better way of averaging the points and seeing the outlier in the bunch.
For something as shown we would either shoot it with a regular instrument or possibly pull offsets and locate those with RTK- looks like that would be pretty easy to do in this scenario.
I'd move out in the open area of that yard and shoot in two random points and at least get a well measured distance with a tape or disto from each random point to the point under cover plus between the random points and / or shoot everything with a TS to verify.
I'd be referring somebody to the unemployment office.
Newer units tend to do better dealing with multipath. They still get bad inits, even in areas less obstructed than the one in the pic.
The other day I took a 20 second shot on a control point. Walked down the street and got a new init with better sky. As I walked by I stopped to get a second shot on the CP. I couldn't get the shot. I used Trimble DL and ran a 20 minute static while I broke out the robot. Did my setup and went to the office (1 1/2 hours away).
The static and robot worked .012 3D. The RTK shot was 2 feet whack. It doesnt happen often but it still happens.
This age old truth still holds- Every occupation and segment of data needs checks at the beginning and end. Your checks should increase as the effort to get the data increases. That wont change.
that always gets me when I see guys that will do all their stakeout with GPS.
I had two people during interviews think we are crazy for not using RTK for layout. You can absolutely do it, but check it again later and see if its the same.
When some pipe slope are 0.17% I don't think so.
I will use it for rough grading and other non-essential items but not for storm, sanitary and curb. Definitely not for buildings - unless rough grade
As mentioned above, it has its place for sure but its not perfect.
Shawn Billings, post: 451927, member: 6521 wrote: Not enough information. Loyal is correct and Jim Frame for that matter. You need multiple fixes spread out over time. Also what are the accuracy requirements? Is this the only point to collect in the area? Is it a small site or large area? I take shots like that almost every day. But I might not if the job requirements could not be efficiently met with rtk.
Typically, our field procedures are as stated above. This particular control point is along the Intracoastal Waterway R/W and the survey was for a new dock, SPC required. The measurements taken the day of the photo did not work, upon returning staking the point was almost 0.75' horizontally and about the same vertically, and in fact the job had to be completely re-done. The lot is not very large, 77'x180' and there is plenty of control in the street where there is open sky. I have no clue why the guys decided this was a good shot. For ambiguities sake we'll call him, "Frank", and "Frank" was known for "miracle shots" under canopies like this. I am sure we will find more blunders in the future from the times he had command of the unit... The kid who took the shot originally no longer works for us though, not fired however, moved off to Gainesville for college.
In this instance, if I absolutely had to have an RTK value, at the least I would throw an extension or two to get up above the canopy. Other wooded areas, I get the guys to "poke a hole" and cut out a hole for the extensions.
If it was a topo shot I wouldn't worry too much about it but someone set a control point there?
As already mentioned, a couple of offsets would have verified the location.
