Exporting RTK Vector Data

Jim Frame

Active Member
This is a repost from the Triumph-LS forum, since it seems like it may be more of a JField issue:

I'd like to be able to export RTK vector data with covariance information from the LS so I can include those vectors in a weighted network adjustment. I'd greatly prefer that the export format conform to an existing standard, and the NGS gfile format (http://www.ngs.noaa.gov/FGCS/BlueBook/pdf/Annex_N.pdf) seems like a logical choice. This format was developed for use in bluebooking; it's used in OPUS extended output, and Star*Net can import a gfile without alteration.

Thanks!
 

Shawn Billings

Shawn Billings
5PLS
That's a great recommendation, Jim. I know that vector export is on the list of things to do, but I'm not sure where it is on the list. I'm also not sure about format, but I think you've identified a great option. Thanks for the contribution and it's good to see you on the board.
 

Jim Frame

Active Member
When is 1.10.3.15226 expected to be available? I currently have 1.10.3.14303, and it shows as current. Needless to say, I'm anxious to try out this feature!
 

Monte King

Member
On the topic of RTK vector data format. Is it possible to export a format that would be recognized by most GPS processing software for adjustment. For instance I am currently using Top(abbreviated as to not use bad words here) processing software and it is capable of importing formats like Top .tvf and Carl .raw or .rw5 files. Sadly the .obn file from the Ashtec days using zxtremes is not supported by my current software. As I am an employee of a fed agency I do not get to make purchase choices(only can offer my opinion) and so I usually have to end up mixing and matching equipment and processing software. The reason I would like this option is that I have become accustomed to the process of setting up my base using auto position and recording the sat. data for post-processing using OPUS or CORS. Then if I record the vector data in a file then later when my base position shifts I can run the adjustment using the vector data and all of the RTK positions shift with the base. With RTN this will become unnecessary however I do end up working in locations with no cell/data connection so I would like the option when needed. By the way I am very impressed with the product thus far! Thank you!
Here is a picture of the LS locating a stone marked and set in 1887 dependent resurvey to perpetuate the original corner established in 1860. The stone is located on the South Boundary of White Earth Indian Reservation which is also a portion of the 10th Standard Parallel of the 5th PM of Minn.
JavadStone1.jpg
 

Jim Frame

Active Member
You can have the LS record data in JPS format, then use the JPS-to-RINEX converter to produce a RINEX file, which Topcon Tools can import. You'll have to figure out how to handle antenna models -- I note that Trimble still hasn't included the Triumph-LS or Triumph-2 models in their latest (September 2014) antenna file, which is a problem for me.

If you want only a vector file with statistical data, the NGS gfile is the only open standard of which I'm aware.
 

Matt Johnson

Well-Known Member
5PLS
The reason I would like this option is that I have become accustomed to the process of setting up my base using auto position and recording the sat. data for post-processing using OPUS or CORS. Then if I record the vector data in a file then later when my base position shifts I can run the adjustment using the vector data and all of the RTK positions shift with the base.
With J-Field you can currently adjust the coordinate of the base and then recalculate the coordinates of all those points collected from that base coordinate.

To do this go to Points and choose a point surveyed from that Base position. Click the edit button and then the button containing the Antenna Height. It will then list the statistics for that point, click the Base button and then enter the adjusted position and click ok. You will be prompted if you want to correct all related surveyed points, choose yes and all your points will be updated using their vectors from the OPUS coordinate you entered.
 

Shawn Billings

Shawn Billings
5PLS
To add to what Matt said, any calculated points or imported points in the database will NOT move with the change in base coordinates. This is reasonable since there is no connection between the base and a calculated point, but it should be understood and considered when using this technique.

Also, when changing the base coordinate, the default system you will be presented for the base coordinate will be WGS84/ITRF2008. You can change this to show any coordinate system in the library and then edit the coordinates.
 

Monte King

Member
Thank You all! Very helpful! I am learning more and more everyday on the capabilities of the LS and Jfield. Great stuff!
 

Matthew D. Sibole

Well-Known Member
5PLS
Has anyone been able to export the NGS G files and process them in a least squares adjustment program? If so what program are you running? I need to find a program that can give me the RPA for the lines on my survey so I can make sure I am meeting min technical standards.
 

Jim Frame

Active Member
I've been using the g-file export successfully with Star*Net v6 (though I can't think of any reason it wouldn't work with subsequent releases). Some hand-editing of the resulting Star*Net *.gps file is required, since J-Field isn't (currently?) able to identify the base point name when it creates the g-file, so it inserts a placeholder base point name (e.g. B001) for each vector record instead. But the editing isn't onerous, and the results have been excellent.
 

Matt Johnson

Well-Known Member
5PLS
Some hand-editing of the resulting Star*Net *.gps file is required, since J-Field isn't (currently?) able to identify the base point name when it creates the g-file, so it inserts a placeholder base point name (e.g. B001) for each vector record instead.
I will add this to the list of things to be done.
 

Shawn Billings

Shawn Billings
5PLS
I will add this to the list of things to be done.
Yes. Now that we require the creation of the base point, it should be easy to implement except for VRS solutions, because the base position isn't stored as a point in that mode. (The base coordinates are stored with the point just not as a point).

Matt, I've successfully adjusted g file vectors in Carlson Surv Net. Columbus is good software. Probably the most robust I've ever seen but not terribly user friendly. Star Net is an industry benchmark, very reputable. If you find yourself mixing a lot of GNSS and terrestrial measurements least squares is the best approach.
 

Jim Frame

Active Member
...except for VRS solutions, because the base position isn't stored as a point in that mode. (The base coordinates are stored with the point just not as a point).
More to the point -- if I understand the way VRS works -- the vector between the VRS base point and the rover isn't, by design, a source of significant error. The thing you'd want to include in the adjustment would be the standard errors of the VRS base point, but I don't know if that information is part of the correction message.
 

Cheyne McLean

New Member
Yes. Now that we require the creation of the base point, it should be easy to implement except for VRS solutions, because the base position isn't stored as a point in that mode. (The base coordinates are stored with the point just not as a point).

Matt, I've successfully adjusted g file vectors in Carlson Surv Net. Columbus is good software. Probably the most robust I've ever seen but not terribly user friendly. Star Net is an industry benchmark, very reputable. If you find yourself mixing a lot of GNSS and terrestrial measurements least squares is the best approach.
Shawn, using Carlson Surv Net, how were you able to import the gfile? I don't see that file type as an option on Columbus or Carlson Surv Net. Thanks
 

Shawn Billings

Shawn Billings
5PLS
Change the file extension of the g-file from J-Field to ".txt". Then in SurvNet select Tools>Convert Total Station/GPS Files. For the GPS/Total Station File Format, select NGS Gfile/OPUS Extended Report .txt.

Once the file is converted to the StarNet GPS vector format (which is what SurvNet uses natively), you'll want to carefully look at the names of the points it creates. If your point names are more than four characters, you'll need to repair them in the text file as the NGS G-File format only uses names with four characters.
 

ndoring

New Member
In the G-File, what do lines G1, G2 and G3 represent? Below is an example of the data once imported into Starnet.

G0 'V90 Day081 17:56 181875gps0-180330.gfile
G1 22-14 211.0109 -262.2626 -200.4688
G2 0.0042 0.0063 0.0059
G3 0.8467573 -0.8574495 -0.8714946
 

Shawn Billings

Shawn Billings
5PLS
From the Carlson SurvNet Help:


The following is a typical vector record in the StarNet ASCII format. GPS vectors typically consist of the 'from' and 'to' point number, the delta X, delta Y, delta Z values from the 'from' and 'to' point, with the XYZ deltas being in the geocentric coordinate system. Additionally the variance/covariance values of the delta XYZ's are included in the vector file.



.GPS WEIGHT COVARIANCE
C PRS34452 1305780.345005 -4667085.299019 4132689.544939 0.005000 0.005000 0.005000 --MON
C UCNJ 1305780.345005 -4667085.299019 4132689.544939 0.00000100 0.00000100 0.00000100 --MON
G1 UCNJ-1000 8399.71318061 -4742.15645068 -8036.07224424 --MNS
G2 3.939428e-006 2.474560e-005 1.160301e-005
G3 3.924536e-006 -3.360765e-006 -1.028503e-006
G1 UCNJ-1001 8328.15569476 -4796.59445569 -8072.25948922 --MNS
G2 9.596618e-007 1.687749e-005 1.936038e-006
G3 1.176891e-007 -8.668020e-009 -4.798408e-006



The first line defines what values are in the G2 and G3 records. It can be either GPS WEIGHT COVARIANCE (G2 is Variance, G3 is Covariance), or GPS WEIGHT STDERRCORRELATION (G2 is standard error, G3 is standard error correlation). This line is optional; the default is COVARIANCE.

The next two lines are coordinate control records. These records are also optional. If used, they must be Geocentric Coordinates (XYZ) in metric units. The format is as follows:

C PointName X Y Z StdErrX StdErrY StdErrZ --PointDescription

The standard errors and point description are optional.

The GO record is a comment.

The G1 record includes the 'from' and 'to' points and the delta X, delta Y, and delta Z in the geocentric coordinate system and an optional description of the rover point.

The G2 record is the variance (or standard error) of X,Y, and Z. The G3 record contains the covariance (or standard error correlation) of XY, ZX, and ZY. Most all GPS vector files contain the same data fields in different formats.
 
Top