The forum has discussed different ways to determine the location of a base station antenna.
This might help.
https://docs.emlid.com/reachrs2/reachview/base-mode/#base-position
For NTRIP it will be specific to your region as to what providers and mount point are available. In the United States HxGN Smartnet is probably to most popular. In Texas I use RTKNET which has been rock solid and appears to work better with our NGS control. Just Google CORS VRS and the Country in which you work.
Hi @himpio,
Could you tell me the different steps to get the position of the antenna (lat, lon, hight)
To choose a method of determining base coordinates, you will need to understand whether your application requires relative or absolute accuracy.
If you need relative accuracy only (for example, for volumetric measurements), you can average base position in Single.
In case the coordinates should be tied to the global coordinates, you will need absolute accuracy. In this case, you should find precise coordinates of the base. To do it, you can use the PPP technique or average base position in Fix using correction from an NTRIP service.
To learn more about different ways to set the base, you can explore our Placing the base guide.
What is CORS method?
Continuously Operating Reference Stations (CORS) constantly gather satellite data to find their coordinates. You can get corrections from CORS via NTRIP to get a precise position of your base. Please note that you will need to request the NTRIP credentials from the CORS operator in your state. To check the nearest CORS, you can explore the map of CORS stations.
What solutions do you recommend for NTRIP caster implementation?
As Michael noticed, available NTRIP services depend on your region. It should be possible to work with a remote NTRIP base in RTK if it meets these requirements:
The distance to the remote base doesn’t exceed 10 km for Reach RS+ and 60 km for Reach RS2
It provides at least RTCM3 messages with an antenna reference point and GPS observables
There are circumstances that you want relative accuracy but need it for 4 - 5 days. This is my situation and that of my customers. If you need the same base for 4 - 5 days, once your average base position has been established, you need to go to the base tab of Reachview and write down the LLH (Lat/Lon/Hgt). Come back the next day, mount the base in the exact same location and manually setup the base with the coordinates you wrote down (you can also do a screenshot for recording the data).
This is how construction does it. The base coordinate doesn’t matter in global accuracy because of the localization.
My business is Agricultural construction. We do it the same way as it is only pertinent to that job so to speak.
I am normally more concerned with the H than the L/L of L/L/H. H is the factor that is used to control automatic grade to put slope on our drainage tile. If I am a bit off on the L/L, I can adjust for that back in the office.
This is the reason I have asked for base station storage in the unit. If you are able to store say up to 5 locations with the ability to download/upload, it would be very nice. We might survey a job in late fall or early spring and install the drainage system in the following fall. We use the same base station location and LLH in the unit as we did for the survey. If we didn’t, the Trimble unit has a Benchmark in it that we could re-bench and make a new base location work but, that can get confusing. We only do it if we absolutely have to. An example was someone stole our metal base stand we made (just a 1" steel pipe, 8’ long, welded to a piece of plate steel to mount the globe on. We sink it 3’ deep in the soil and leave it till we come back. Apparently, someone needed it worse than us.
In a localization H is irrelevant as well. Shoot in one known elevation (preferably two on opposite sides of the site) and use them for H only. Then use the property for LL only.
I’m surprised no one has mentioned OPUS – the Online Positioning User Service.
OPUS can help fix the absolute (real) location of a stand-alone (uncorrected) L1/L2 base station (such as the RS2) to within a few centimeters with observations of 15mins to >24 hours depending on the location & quality of GNSS reception:
https://www.ngs.noaa.gov/OPUS/about.jsp#about
Note: you do need an internet connection of some kind in order to submit (upload) raw GNSS / RINEX logs to the OPUS servers. This is a post-processing step that can be done any time, it does not require a real-time internet connection.
H is very relevant in my circumstance. H sets the benchmark which is what is referenced for all the tile we put in. We design the job with Trimble software then go back and install. If H is off, so is the depth of our tile which can be very critical. We strive for a depth of 2 to 2 1/2 feet deep. We are shallow due to heavy clay soils. Our spacing is normally between 20 and 30 feet
H as far as global accuracy is important? The site is relative to itself so the site benchmark could be 500’ or 1024.78’. It shouldn’t matter.
If H is off from the original base LLH, our depth is off equal to the difference in H. We can go to our original Benchmark and re-bench which corrects for the difference in LLH but if you forget to, your first run is messed up. When plowing in drainage tile, there is no correcting a bad run other than installing a new one to the side of it.
The system sets a master bench (Which references LLH). Everything is converted into XYZ from the benchmark with the benchmark being 0/0/0. You can create others (BM1, BM2, BM3 etc.) as you go but it is a hassle and gets people confused. If you have the same relative LLH/base location, it isn’t needed.
When the original base is created lets say I get a 42.780501/-84.145553/285m, If I create a new base (which is normally done under WAAS data), lets say I got 42.780XXX/-84.145XXX/282m which is say 10 feet off from the original LL and the H is off by 3m it would install my tile 3 m deeper.
A re-bench is done by going to original physical location of the benchmark (not the one indicated by the monitor as it is off by the 10 foot difference in the LL) with your GPS globe directly over the same point you tap the re-bench button and it brings everything on the same playing field.
The monitor takes into account your globe height difference from your survey vehicle verses your machine that you are installing with. This is all done in the initial setup of your vehicle or implement.
In another thread where we were talking about base station correction rate (Hz) I think it was Polina that said 1 Hz is sufficient for survey work. This is true for typical survey work where you take a reading for a set amount of time. We survey with a John Deere Gator normally at about 20 MPH, that is why 5 HZ is preferred. The Trimble system is taking a LLH reading every so many feet (which is user determined in the monitor, normally 10’). Depending on the topography of the ground, we are doing swaths ever 10 to 50 feet. Flat ground every 50’ is sufficient, rolling ground needs closer spacing in order to do a good topo map. The software is basically averaging data between the point data to determine the topography.
Perhaps this weekend, I will start a new thread with a start to finish description of what we do.
Doesn’t manual coordinate entry take care of that? If you shoot future point off that then everything is still relative. Like you said, base station configuration saving is a must!
This is a major reason why I want localization in Reachview. You set the base in one spot, get an instant (1min) I am here coordinate and done. Locate any site control and then you are relative to that network regardless of the starting point. The base is not even used H or XY as part of the localization adjustment. Just make sure the base is always on that setup. This is also why we use permanent base poles or at least a rover rod with a tripod attachment.
We usually run 10Hz on our equipment and my truck. This accounts for the range of 5-10ft/sec.
If the base LLH is the same the Benchmark location is the same, if not, no we are off as it still references the base LLH.
5 Hz has done a good job with 3rd party GPS receivers, some do not have the 10 Hz capability. I haven’t dug into ft/sec.
Trimble’s Construction Division can do things our Agricultural Division cannot, moving base is one example. Perhaps what you are speaking of is another.
I think I am not getting across the idea of localization which you can do in FieldGenius. If you have a benchmark to set the localization elevation to then even better. I now recall that you are automating equipment right? If so, how are you inputting the data into the machine?
FPS is a good way to calc either you speed or the interval of the shots. It’s more important on the machines as the are controlling the hydraulics on a continuous stream instead of “taking a shot” every 20ft or whatever. This is especially important on a motorgrader because they can easily hit 10mph on finish passes so at 10Hz they are getting fed a correction every foot.
The Trimble monitor is constantly correcting for any movement over what you set your deadband at, unless you are not moving. If you are not moving, it has to be greater than 1", this way the machine isn’t constantly seeking while sitting still. We normally have our deadband set at +/- 1/4" So if elevation or height display anything over 1/4" it sends a correction to a solenoid valve either up or down to bring it back into tolerance.
Unfortunately, our benchmark doesn’t work the same as what you are explaining
Trimble doesn’t do anything than any other machine control solution does so nothing odd there. I’ve managed Komatsu, Trimble and Topcon and the internals in most of the sensors are the exact same boards. As a matter of fact the Komatsu are Topcon with a gray shell. Topcon auto stops all action when the machine stops.
We’re splitting hairs because we are talking about 5 vs 10 corrections/second, but according to grade check with MM-GPS it does make a difference on a blade at speed. Most notably on slopes. Considering the IMU’s action is 1000Hz the control box will take whatever you can send.
What most people call localization Trimble calls Site Calibration and you must be running with some customizations because as far as surveying goes you have to calibrate. This is why there are rotations and scale factors in the CAD we use. Those localized control files then carry throughout the process as part of the project file.
On the Ag side of Trimble, there is no Site Calibration. I am guessing you were or are working with Trimble Construction Equipment. The only thing we had was the original WM-Survey which used the construction side software to do survey work. When it came to Ag they took out half or better of the components.
The software for Ag used to be called Farm Works, it has since changed names to Trimble Ag Business solutions (TABS) and has since became a cloud only type software other than the WM-Subsurface software which we use for tile designs.
The only way to make the pre-surveyed data match at a later date is either using the exact same base location with the same LLH or doing a rebench.
Now I understand. I’m not an Ag guy, but it is odd to me because localization would be the one thing out of what we do that I would think is perfect for maintaining a field. All you need are some fence posts, a topo, a way to define the new topo and a machine to do it. Regardless of actual elevation. Ag seems to have branched off since they started it.
Here is the begining of a survey file by the FMX or TMX monitor, I took out the smaller numbers and replaced them with X’s
<?xml version="1.0"?>-<field_level>
- This is the Benchmark Location relevant to the base
<gps_offset>
0.000
0.000
0.000
</gps_offset>
-
0.00000000
0.00000000
0.00000000
I took out some of the blank lines to reduce space.
Basically, all the points are referenced from the BM i.e. distance from the BM and sea level ALT
Once this data is collected for both the boundary and interior it can be brought into the software. I have attached a cheat sheet that I made for my customers that briefly describes what and how I layout the tile designFarm Works Surface cheat sheet.pdf (3.2 MB)
