GPS is a STATISTICAL measurement tool. (Think 500 gallon tank, and a 1 ounce sample.) Meaning, it uses probability. IF you set up your whole system, and let it collect for 24 hrs, then post processed, and cleaned your data, you could get real close to the truth.
Each shot is a LITTLE sample over a little period of time. Say, 3 minutes. When you work in deep woods, you add more error per shot, or, it's like taking that little 1 ounce sample, BEFORE thoroughly mixing the 500 gallon tank.
In this post, I want to show several screen shots, taken from a recent survey.
These screen shots are turned on, in my GPS system, in case of the need for later evaluation. So, they are available for EVERY point taken by me for 5 yrs.
?ÿIn this First Pic, we have Point Number 4813. The LEFT hand square shows HORIZONTAL shot spread. Kind of like shooting at a target with a shotgun. the Right hand square shows the vertical shot spread.?ÿ The Star shows the RTPK shot being in agreement From RTK to RTPK is: (N-.003' E-.013' and Down 0.125'). The size of the left hand box is 0.203 feet, and the size of the Vertical box is 0.266 feet. This shot took a total of 6 Minutes. It is in deep dark woods. What this is saying, is that it is confident that I have that shot within 0.203 feet.
I want it more accurate than that. So, I re-shoot it.
Here is the second shot. 4814. It generates an even larger left hand square on the left, (Horizontal). It's 0.25'. Also, RTPK and RTK disagree now by N-0.084' E-0.073' and Down 0.087'.
And, the difference between these shots is S26?ø48'W 0.18', (This is from Point Number 4814 to 4813) This is shown on the box 1 up, and 1 left from the SE corner of the screen shot.
Now, this is too large of a difference for this shot.
So, I shoot it a 3rd time.
Here is the 3rd shot.
Here, we get again a large shot spread, (Left Box is 0.25' and right box is 0.391)
However, the inverse between 4815, and 4813 is S02?ø23'51" W 0.062'. At this point, I decide to shoot it one more time.
Here is the shot:
Here is the 4th shot. Now, the box on the left is much smaller, (0.078' Horiz. and 0.168 vert)
Inverse from the first, to this 4th shot is N 27?ø49'27"W 0.033'
I now use an average command, and it makes a coordinate that is nearly on top of 4816. It is called 4813A.
This is my final coord.
The 4 shots have taken my 11 minutes.
And, I'm done.
I did not take a pic of the area. But, it was 60' hardwoods and pines. It's only 5 miles from my house. If somebody asks, I'll take pics of the area.
I felt that my happiness with the Javad system deserved an explanation.
Now, go get ready for new year!!
?ÿ
Nate
?ÿ
?ÿ
?ÿ
?ÿ
?ÿ
I
?ÿ
?ÿ
Thanks Nate!
?ÿ
So,
When are you officially going to start repping and selling the Javad gear anyway?
????
And also provide some independent (i.e. non-GNSS) proof of said (unicorn?) results. A control traverse fully adjusted or certified baseline in the same conditions would appear the relevant comparison. GNSS in open vs total station results not in dispute just the canopy/multi-path stuff.
?ÿ
Providing an output from a single level, total station, whatever without proof (initial calibration, regular checks) is meaningless. Can't see how providing multiple observations from any single GNSS is any better.
@lukenz?ÿ
Lighten up Francis....
???? ?????ÿ
?ÿ
https://m.youtube.com/watch?v=iN-aXzpQUdw
?ÿ
Happy New year!
?ÿ
?ÿ
This reminds me of a story. My dad was hired by a Client named Clay H Logan. Dad, and Brother ran a traverse around a big wooded lot, with a HUGE brush pile on it. They finished the survey. Clay called them back. He said, "I'm payin you to run the TRUE line, right through that Lot, and over the top of the brush pile. If your work lines up, I'll pay you. IF not, I am paying you only for running the true line". Well, it was a big brush pile. Dad said, ok, it's your money. However, we cannot get over the pile very well. How about IF we run the line from BOTH ends, and you get up on the pile, to see if they line up? He said OK.
Another day of work. They ran true line. right up to the pile. Then, from the other side, right up to the pile. As he climbed down from the pile, he was said to have muttered "Alchemy".
He paid for both days.
It's called science. Yes, tolerances loosen up a little, in canopy, and you have to go slower,?ÿ but MOST of the time, we are talking about a tenth. And, if you care about it better, you can get it tighter than that.
Now, don't forget to write 2022 on that check! (Not today, but soon!)
Nate
?ÿ
?ÿ
What does the 27 +6 mean in the satellite square in the upper left of the screen??ÿ 27 satellites available according to the almanac, and it is locked on 6 of them?
?ÿ
and what is 782.1 mi away?
27+6 means it's using 27 satellites, and tracking an additional 6.
That screen is the "stake to line" screen. So, it has the left/right and station. It also has an arrow pointing which way, and bearing.
The DTT button is based on an internal magnetic compass. It's the distance to the intersection based on the way the LS is pointing.
You choose what buttons you want on the screen, for various functions.
Nate
@Bill93, if you hit the "review" button, then you can see a graphical view of points, lines, etc. That's what scale it's at.
It zooms in pretty fast. Press the button in the se corner, and it zooms in, press the button in the SW corner, and it zooms out. Slide, and it pans. I didn't post any graphical screens, as I usually don't need them. But, they can be handy.
N
We have a Javed base/rover setup. I found that once you get past the fact it looks different, it works well. I do have issues with the way many people use them, but yoj can't blame that on the tool.
We have done comparisons between our Javad and terrestrial data. I've found two things.?ÿ
First, the data quality reporting is similar to the other brands I've used. It is nearly always overly optimistic by a factor of 2 to 2.5 times. It does have other tools where you can get moee realistic (and often superior) results.
Second, the supporting gear looks real cool, but it is not nearly robust enough. The telescoping rod is absolute crap. The metal detector is nominally equivalent to a cheap pin-pointer.
Bottom line is, the recievers are step up from almost everything out there as far as data. The software is much more intuitive than I expected. Get solid tripods, rods, etc. and you'll be fine..
Happy New Year..
The Star shows the RTPK shot being in agreement From RTK to RTPK is: (N-.003' E-.013' and Down 0.125').
I had to look up what RTPK is.
TRIUMPH-LS combines the strengths of RTK and PPK into a system that can post-process your RTK data and verify its results in parallel and real time.
That's amazing!
?ÿ
thanks for the response.?ÿ 27 is a lot of satellites for the woods.?ÿ I'm guessing it tracks all GNSS constellations??ÿ
The read-out on the bottom, is that state plane coordinates??ÿ Or is it some other projection??ÿ
@Joe B
Your questions deserve a longer answer, than I'm willing to type on this phone. So, I'll answer you properly a little later.
For now, I think I'm running only 4 constellations. But more are available.
And, I'm pretty sure that the coord at the bottom is SPC.
But, what's interesting is you can have multiple local coords all with different. Scales, rotations etc.
Since the internal coord is lat/long, you can express multiple local systems. Without changing the actual lat long. I'll have to do this later.?ÿ
Thanks,
Nate
Thanks, I think I understand. It is a lot of satellites for deep in the woods.
I think GNSS is usually in geocentric X,Y,Z natively.?ÿ That is pretty cool if it can display multiple coordinate systems on the same job.?ÿ I'm not sure if the Leica can do that.
When I look at the location of the coordinates it takes me to a spot right along state highway 240.?ÿ is that right??ÿ Are you sure that is the snapshot to the point you are thinking it is??ÿ I just expected it to be deeper in the woods.?ÿ Maybe you uploaded the wrong screenshot?
For me, I always do 2 sessions on a point if I want to make sure it is correct.?ÿ doing observations within 10 or 15 minutes of each other doesn't really give you a separate solution, because it is all the same satellite geometry.?ÿ I think Javad is pretty well known for its robust multipath mitigation.?ÿ If you're getting 27 satellites, there shouldn't be an issue though.?ÿ Looks like a cool unit.
But, how does it do when it's so deep in the woods it only gets 6 or 7 satellites??ÿ Don't tell me you always get around 27 sats??ÿ That would be pretty amazing.
?ÿ
25+sats used is normal in deep dark woods.
Still replying with thumb, on phone.
That coord is close to correct, within 100' . It's near Hwy 240. It's 94' south of the center section. Staking line towards the s14 cor.
It's an old bastard coord system, from yrs ago.
Nate
OK, the coord at the bottom of the screen, is Arkansas South Zone, SPC, NAD 1983 2010 epoch date.
So, it does not express the local coord. It is the State Plane one. I wanted to be sure.
Nate
?ÿ
?ÿ
we always called persons a transit head if can stand in the middle of a long line and say our work is wrong. you turn a tight double angle and they still aren't satisfied. just this past week had one of those. i won't say what i really call those persons, usually just hey bo, go take the exam and work 30 years in the woods.
@Joe B
"For me, I always do 2 sessions on a point if I want to make sure it is correct. doing observations within 10 or 15 minutes of each other doesn't really give you a separate solution, because it is all the same satellite geometry"
I would disagree. The satellite constellation changes significantly in only a few minutes of time. They are moving pretty fast. I use 4 minutes to check my shots. The receiver I am currently using will give the same bad fix within my 4 minute window, but every other receiver I have tested will not. Longer time between shots is obviously better, and I recommend developing test procedures that you know are good before you rely on a shot.
I have tested my procedure with total station ties; other surveyor's record data; opus-rs solutions; and anything else I could think of.
I just want to clarify that The scenario I'm thinking of is in harsh conditions, without doing any other checks to the point, I would do 2 sessions.?ÿ For instance, if I have a 50-acre property, and I GPS a corner, in poor conditions, with minimal satellites, and no total station or anything else, I would need to do a 2nd session.?ÿ Even if I do multiple observations within 5 or 10 minutes.?ÿ I will still go back to do a 2nd session, because the risk of something being out, even if it is unlikely, is not something I would risk.?ÿ I believe you because I do the same checks over and over.?ÿ It's not like the work I'm doing is out by multiple feet, and I have to correct it all the time.?ÿ It's that the risk of it being out, which I've seen happen before, is not worth it over going back and checking it again.?ÿ And I charge for that confidence and commitment of my time. I'm not saying anyone is doing anything wrong if you don't do it my way...?ÿ Just to be clear.?ÿ There are many ways to skin a cat as they say.
As for the constellation changing significantly in 4 minutes, I have to disagree with you.?ÿ I agree that satellites are moving very fast.?ÿ They are moving at roughly 7,000 miles an hour.?ÿ At 12,000 miles from the earth.?ÿ 7000/60 = 116.7.?ÿ 116.7*4 = 466.7 miles in 4 minutes.?ÿ Given the distance of roughly 12,000 miles from the rover's location on the earth, that is an angular change of 2.2 degrees of change.?ÿ That isn't a huge difference, in my opinion.?ÿ The signal is still going through generally the same obstructions.?ÿ?ÿ
Also, all the satellites are moving at roughly the same speed and.?ÿ I'm sure there are slight variations in the geometry of the satellites, but I don't think it would be significant.?ÿ?ÿ
But, if you have, say, 15 or more satellites, I can see why you would be confident in that.?ÿ But if you're in a really bad place, and only have 6 satellites, I would not just trust anything the GPS unit has to say and would go back for a 2nd session if I needed to be confident in that position.
I'm sure part of it has to do with where you are located.?ÿ Where I live, it is not a good place for GNSS.?ÿ You can't do a job without breaking out the total station.?ÿ So, that might play into my opinions on the unreliability of GNSS.?ÿ?ÿ
Mathematically you are correct; but if you watch the satellites dropping in and out of the solution over 4 minutes you can see what I mean. I usually have 20 or more satellites in a solution, so my canopy isn't likely as dense as yours. I rarely work in Nate's part of the State. I would likely have to break out the total station if I did. Sometimes it takes an hour or more to get a reliable set of shots in canopy. I always regret staying on a point that long. You can do a lot of traversing in an hour. It usually just happens without any forethought.
The issue with multipath is that the signal gets bounced off objects and the distance to the satellite is changed. I've never seen two separate fixed solutions be off the same amount. From the days of daily bad fixes I always noticed the distance to the incorrect fix was around 6-8" horizontally. The vertical was always more "off" than the horizontal.?ÿ
The chances of two identical fixes both being off the same amount in the same direction depends on getting a fix with the exact same multipath conditions.?ÿ
Letting the satellites move will negate that, but the simpler way is to move the receiver. Raise or lower the rod and the multipath the receiver is experiencing will change. At that point two fixes with the same result puts the chances of a bad fix in canopy into the statistical category of finding a unicorn. Of course accuracy is another issue, if you need less than 8mm you need time and clear skies.?ÿ
?ÿ
?ÿ
?ÿ
?ÿ
?ÿ
?ÿ
?ÿ
?ÿ