I am setting up a Topcon HiPer V rover for a customer in New York. I left his receiver running on external power connected to his data collector which was also on the charger.
As I was sitting at my desk working on something else this morning, the voice annunciator on the receiver said "RTK fixed."
When I looked at the status on the receiver it said the receiver is 614 miles from the reference point. RMS values were in the 1.5 foot range.
That's cool!
They are all getting better.
Nate
NYSNet Salamanca Station??? A 614 mile radius around Shelbyville, TN, lands right in the middle of Salamanca where there is a CORS base running...
nice Brian
1,197.020 miles, post processing CORS data. I wanted to see two things:
- How well independent solutions agreed;
- How close the position of a CORS site calculated from other CORS throughout the country would be to the published values. Pretty darned close as it turned out. All the baselines longer than this one returned float solutions, but they all still fit very well.
Note the precision estimates at 95% confidence as well as the 0.055' difference in vector length. I used data from two completely different days and times to compute this.
The other attached photo is the network - I held CORS in AZ, ID, NE, OH, NC, FL, and TX to compute a point in LA, and I kept and adjusted all baselines (note the scale bar bottom left). The deltas between the calculated point and the published point were north 0.013', east 0.037' and elevation 0.127'. Pretty amazing, IMHO.
Max distance to the base receiver fixed integer: 550 nm. Which is just over 1,000,000 m.
My experience has been that an accurate integer fix remains valid at long distances. Base receiver was only 6 km away, not a CORS. Flight was generally 4.5 hrs. Zero significant cycle slips. The solution returned fixed integer throughout. (In Waypoint's Grafnav there is an option to not drop integers based on distance.)
Max distance to the base receiver: 550 nm.
So, a single base receiver, close star/finish, fixed integer, forward/reverse solution. RMS <10 cm. (1/2 carrier wavelength)
As always, a 2nd and 3rd solution was also generated using NRC-PPP, and NASA's Gipsy programs. All three solutions are compared by subtraction. Differences were 5-8 cm, after removing 5 cm bias in height. (The control point position was not on the current GPS epoch.) A the 5-8 cm diff was not distance dependent; about the same beginning, middle, and end.
I've been told this can't be; at a million meters it's a couple of meters, best case. And the flight profile was always 40,000 feet altitude and 400-450 knots. However, I've seen this, repeatable performance, dozens and dozens of times, at 200-400 nm. And also with multiple base receivers 200 miles apart, I've seen very similar comparisons of fixed integer, long distance, compared to each other, and to Gypsi and NRC-PPP. Granted, not 1-3 cm, generally 8-12 cm (1/2 wave). Now that's 10 cm, not the hair splitting 1-2 cm requirement for Lidar, that's a very different animal. Navcom's Starfire solution (realtime) was not much different. Just a slightly larger vertical bias.
It was very interesting. Not a one-off. And 4+ hrs of continuous, slip free data req'd for a PPP solution to check the base receiver solution.
A long time ago at a CLSA conference a paper was presented about long radius fixed integer kinematic solutions. And he discussed 200 miles, single frequency, airborne data. But that was 1991.
During the government shutdown in 2013 I sent data collected in Oklahoma into AUSPOS which returned a position based on CORS on the east coast and in Canada.
Well, a guy at a company I was working with was doing topo while I was looking for corners. He said he was having a terrible time keeping his Rover fixed RTK,,,,,,,,turns out he was getting a radio from a base 60 miles east,,,,,,needless to say he had to redo the topo,,,,,he was fixed the whole time.
Wow, I thought 40km was about max and figured it just would take forever to compute a fix beyond that (So I haven't tried). I'm tempted to register for Vermont's RTK network and see if I can obtain a fix 44 km (27 mi) away.
MarkO ...
That's not radio linked RTK. That's phase differential with a base receiver, post processed. And integer fix determined while stationary at 6 km to a base receiver, not a radio link.
If there was a serious cycle slip at even 50 miles out, at 400 knots, the solution would remain float.
JerryS, post: 369858, member: 205 wrote: I am setting up a Topcon HiPer V rover for a customer in New York. I left his receiver running on external power connected to his data collector which was also on the charger.
As I was sitting at my desk working on something else this morning, the voice annunciator on the receiver said "RTK fixed."
When I looked at the status on the receiver it said the receiver is 614 miles from the reference point. RMS values were in the 1.5 foot range.
Wait, it's only 600 miles to New York?
Wasn't aware of how post processing worked. I figured it compiled the data and calculated an occupation. Are you saying it uses phase shift to calculate a fixed position similar to RTK?
Here in MA, I no longer use base/rover now that we have an RTK network (and station barely a mile from my home). I've tried the MA DOT network just over the border in NH and also in RI and it computes a fix pretty easily. No luck at my cabin 90 miles north. But I know VT has an RTK about 25 miles to my west, just figured it was too far for practical uses. Thoughts?
RTK is a local solution, and not the more rigorous phase differential post process solution of precise work.
Mark after Katrina I had people shoot RTK shots across Lake Pontchartrain from stations in Jefferson Parish, which is a distance comparable to the 27 miles you mentioned. It wasn't using a radio, it was using IP connections. The results were hit or miss; the verticals would be good if the conditions were right, but sometimes they were 0.5' or more off as well.
I think the trick to getting a good RTK fix at that distance (along with clear sky, low multipath, good DOP, plenty of SVs, yada yada) is that the weather has to be the same at the base as it is at the rover.
Well my baseline distance record of 94km or so is a bit short, but I reckon it could be up there in terms of radio broadcast RTK. 25W or 35W base radio.
Larry Scott, post: 369997, member: 8766 wrote: RTK is a local solution, and not the more rigorous phase differential post process solution of precise work.
Hello Larry, could you please expand on what the 'local solution' of RTK is, as opposed to a phase differential.
As a postscript of sorts, I switched data collector software to Topcon Magnet software with the same set up otherwise and it said it was fixed in about two epochs.
Mind you, I am not suggesting that this is something anyone should try other than as a point of interest. I was merely setting up a system to work with the customer's circumstance. I never expected it to say that it was "fixed" at that range.
And as a minor correction, the baseline length to the mount point for the iMAX solution was 613.1 miles.
As the crow flies, Tommy Young, that 613.1 miles does not get you to anywhere in New York so I know there is more here than meets the eye.
Sorry for the newbie question, but what is an "epoch"? Is it the time period for a satellite to orbit the earth? I know it has something to do with time, lol.Also, FWIW, Shelbyville is less than 600 miles to the western portions of NY state, (per google earth) so maybe...
Showoff...
What's the longest EDM distance you've ever shot?