Out of curiosity, how many folks are actually doing their own Post Processing into the NGS CORS Network, in order to align themselves with the NSRS?
This would NOT include those who use OPUS (Static, Rapid_Static, Projects).
Of particular interest (to me), is just how you go about it...
What software do you us?
What CORS Coordinate estimates do you use?
What ephemeris do you use?
Which antenna calibration models do you use?
The reason that I ask, is because I recently received some Static GPS data that was computed by someone who had NO CLUE about how this SHOULD be done.
Loyal
I don't know of anyone doing it.. What is the advantage?
I do it all the time. Don't have time now to go into details. I will try to do so later.
Mike
> What software do you use? leica geo office, versions six or eight, whichever license is available
>
> What CORS Coordinate estimates do you use? NAD 83(2011)epoch2010, ngs published
>
> What ephemeris do you use?usually igs rapid, occaisional precise
>
> Which antenna calibration models do you use?NGS absolute pcv's, vintage 2011
>
hi loyal, hope you are well. AND i hope i passed your test.
sincerely,
'grasshoppa'
I Have Done It
Most often when OPUS says my favorite CORS data is not yet available or is not up to par.
I download CORS RINEX and my Ashtech processors comes up with tight Solutions. I use 3 CORS holding each one fixed and recording my receiver position. I then mean and hold my receiver fixed recording the control ties to the CORS. A couple of days later I submit to OPUS again, check against my solution and put the OPUS report in the project folder. I have yet to see anything that required me to change my own solution.
I had a lot of experience processing L1 only data brfore I ot L2 receivers. since I still use a mix of receivers and am almost always close to at least 1 CORS, I just automatically put the CORS RINEX into my solution mix.
More often than not I never use OPUS, except as a quality check after I download. Instead I break up longer files and submit multiple observations to OPUS-RS. I am much more pleased with the precision.
Heck I seldom even wait for the Rapid orbit data, Ultra-rapid being just fine.
Paul in PA
> Out of curiosity, how many folks are actually doing their own Post Processing into the NGS CORS Network, in order to align themselves with the NSRS?
>
> This would NOT include those who use OPUS (Static, Rapid_Static, Projects).
>
> Of particular interest (to me), is just how you go about it...
>
> What software do you us?
>
> What CORS Coordinate estimates do you use?
>
> What ephemeris do you use?
>
> Which antenna calibration models do you use?
>
> The reason that I ask, is because I recently received some Static GPS data that was computed by someone who had NO CLUE about how this SHOULD be done.
>
> Loyal
Loyal,
I do it on a regular basis. I'm using LGO with absolute antenna value. I try to use the precise ephemeris whenever possible but will use the rapid if need be. As for coordinates, it depends on what we are doing. The proper answer would be to use the record coordinate with the appropriate sift applied to bring it to the current epoch.
As for those whom questions the need for post processing your own data. I do it to tie multiple points and observations into a network. This could be done with use of projects but I've got the software and have been post processing for, and this really hurts to say, decades.
Merry Xmas to all
John
> Out of curiosity, how many folks are actually doing their own Post Processing into the NGS CORS Network...
I've done it, using TGO/StarNet and LGO. But not very often. I'm already doing more than 95% of my competitors by attempting an adjustment at all.
When I do it I download the datasheets and best available ephemeris along with the CORS data from the User-friendly CORS site and use that.
I always used the HARN.
The CORS were always just too far away for tight results.
Not often, BUT if you want to do this correctly, you basically need to follow ALL the steps that OPUS goes through! i.e.: you should process using the IGS08 positions for the CORS, the new absolute antenna corrections and some flavor of post-fit satellite orbits. You can then use HTDP to roll the position to NAD83 at whatever epoch you want.
Do I always do ALL of that? No! You are mixing and matching data from different datum's if you don't though and introducing some unneeded errors into you solution.
I have found with OPUS on NAD83(2011) that I can get a multiple day solution that has error ellipses well under 1 cm, AND will compare very favorably to published positions at the 1-2 cm level, so for about 99% of my work that is good enough AND quite a bit less labor intensive than rolling my own.
GNSS/NGS/OPUS has made it possible for the ordinary dirt surveyor who uses proper care and technique to position points far better than the best geodetic surveyor could of a generation ago!
SHG
I do it on every job that deserves to be tied to Grid .
We are lucky enough to have CORS stations in the three cities that I work in .
I use Topcon tools to post process my Static work . In Topcon Tools It will automatically search for the Cors stations by Radius and then ask you to import the ones the you want to use . It automatically gets the time and date of the occupation and imports the file to use . I use this file as the fixed control and then process my base only or if I am close enough I will process the hole file in the mix . So I have two vectors into each point . I did a job last week with 18 static points all with a 10 minute observation (30 acres site) and processed with my local CORS and base and all but two passed. So I just took out the bad CORS vectors to and reprocessed .
I also compare the my Cors solution to OPUS on every job as a check . Horizontal is usually a couple hunderths , vertical is around a .08' .
Typical boundary survey:
Set up a Base Station on an autonomous position and go measure the controlling monuments with the Rover. After the controlling monuments have been tied, I will set up the base in a different location and repeat the process; typically the next day and I will try and do it in reverse order to obtain different constellation geometry.
Once I have all of the GNSS occupations in the bag, I will import the data directly from the receivers into a file on my PC. From there, I will import into the processing software. I have been using TBC v2.70, but recently I purchased two Topcon GR5’s and I am learning MAGNET v1.1.1. After all of my data is imported, I look at the occupations and make sure all of the station names, heights and antennae models are correct. I will then import RINEX data from the CGPS/CORS. In my case, I use mostly PBO (UNAVCO) CGPS. I compare the import data to the site logs and verify that the station names, heights and antennae models (NGS absolute) are correct. I will ultimately use rapid or final ephemerides, but I will go ahead and import what ever is available for the first go around. I fix a coordinate (IGS08) and process all baselines. After processing is completed, I will weed out vectors that don’t pass the QC criteria (loop closures, etc.) or have flat out failed. I will process again and export a vector file.
The adjustment is run in Star*Net v6.0. First I import the vector file. I then run a minimally constrained adjustment. I do not solve for scale, rotations, geoid or deflections at this time. I save the adjustment report as “minimally constrained”. I then scale the errors and begin fixing CGPS one at a time until all or partially all are fixed, observing the adjusted free stations coordinates, error factor and sum of the residuals as I go. Once I have fixed all I am going to fix, I run a final adjustment, solving for scale, rotations, geoid and deflection modeling. I don’t bother attempting any horizontal time dependent positioning unless I am trying to get to a particular older epoch. I try not to mix and match, it usually blows up the network.
I have been comparing TBC to MAGNET and have found that the two will ultimately derive the same positions but the 95% confidence region for the unknowns have larger error ellipses after being run in MAGNET.
Closely related question.
I am a GPS post-processing novice. Need a sense of magnitude check.
Just curious, if you post-process in terms of one RefSta.
Then, post-process in terms of another RefSta.
(one hour occupation, 5 sec interval)
(RefSta are say within 20 mi)
Assume that both RefSta are say CORS.
What kind of mismatch in result is "typical"?
Assume that both RefSta are from two different systems.
Say one CORS and one UNAVCO.
What kind of mismatch in result is "typical"?
If there is a "significant" difference and you are using the
exact same observational data is the difference an indication
of the accuracy of the RefSta or measurements?
ex. the vector is too short so solution is pulled towards each RefSta.
(Trimble Pathfinder Office, TerraSync, Pathfinder Pro L1 only)
(Used "built-in" RefSta DataBase, all reduced to same basis)
PS If this should be a separate Thread, I'm sorry.
I may submit other educational questions in future.
I used to do it often with SKI Pro, but since I did the training for OPUS Projects, I have been submitting all my data to it.
With PM3 L1 only receivers, that is all I do.
Either download info from CORS site or more frequently from TxDot sites as they are more and closer.
Nearest CORS is 50mi away. Nearest TxDot is less than 10.
On occasion I can use several at a time and sometimes throw in a CORS site too if I don't mind taking the time to download all the different time sequences to zip and transfer.
:coffee:
A Harris, Why Do You Download ?
With GNSS Solutions, especially using VRS, you can set up automatic downloads.
Aren't the TXDOTs CORS?
Paul in PA
A Harris, Why Do You Download ?
Not all TxDot sites are certified CORS and can easily differ 2m from OPUS solution
I rarely need to know exactly where I am on earth.
I need my networks and individual projects tied together and to locate far away monuments.
You would have to visit the TxDot site to understand the need for manual download.
Strictly as a redneck science experiment I cheat and send the CORS files to OPUS-RS. Then I process our one receiver in topcon tools against those CORS files and the OPUS-derived coordinates for them.
Sometimes this is 24h after the fieldwork and uses the rapid ephemeris, but before the job is done it gets redone with the precise. Eventually I check them against published or maybe time series and it's always "good enough" for the purposes.
Usually this is for stuff that is never going to be reported in terms of geolocation anyway. But it could be, and it fits. Usually also tied with direction sets and trig levelling, so if the GPS positions are at odds with gravity or line of sight, it shows up in the least squares.
Thanks for the comments, and insights into how different folks trying to do essentially the same thing.
Loyal
I do all the time with an expedited process using PAGES-NT, and Columbus Best-fit software. I also use the latest version of HTDP. In the past I've only used fixed CORS positions without standard deviations.
Since the latest version of ADJUST is implementing CORS positional standard deviations, I will try and duplicate the adjustment process using constrained coordinates. The latest ADJUST guidelines is a good read on the subject.
I don't think TBC has kept up with absolute antenna calibrations for non-Trimble antennae, so I've all but given up on CORS processing in TBC.
IS
> Out of curiosity, how many folks are actually doing their own Post Processing into the NGS CORS Network, in order to align themselves with the NSRS?
>
> This would NOT include those who use OPUS (Static, Rapid_Static, Projects).
>
> Of particular interest (to me), is just how you go about it...
>
> What software do you us?
>
> What CORS Coordinate estimates do you use?
>
> What ephemeris do you use?
>
> Which antenna calibration models do you use?
>
> The reason that I ask, is because I recently received some Static GPS data that was computed by someone who had NO CLUE about how this SHOULD be done.
>
> Loyal
Loyal,
Thanks for posting this question. I am using Trimble Business Center (2.81) for post processing, using OPUS as a check. I am rather new to this, so still learning. As for CORS coordinate estimates, I use the "internet download" function within TBC to download data from several nearby CORS stations, all more than 20 miles away. After processing baselines, TBC will flag baselines that don't meet certain acceptance criteria (so far, I've just used the default values provided by Trimble), and as a result I will remove CORS points if it appears they are suspect, and re-process. As for ephemeris, I will download the best available IGS orbit. Typically, Rapid is available at the time I'm processing. As for antenna calibration models, antenna model is set to "Automatic" and antenna model is set to either NGS, IGS or Trimble as determined by TBC software.
Al
