Converting Reach points to local state plane grid

I’m trying to use a Reach RS+ for GCPs at a mine in the U.S. that is using its own local grid coordinates.
Can anyone explain how to convert a point taken with the Emlid into their local (state plane) system?

you can do it the simple low accuracy GIS way and open the survey file in qgis and save it into the state plane projection (with zone) you desire. or you can do it the surveyor/geodetic way and research what datum your coming from and what one you want to go to and then find the tool to do that. unless your base station control is really accurate you might as well do the simple way.

as for the local grid system, there are others on this forum that do that kind of stuff and might chime in.

I would like to be as accurate as possible. I have the global coordinates of the base and those coordinates converted to the local system using the mine surveyors’ Trimble.

I can convert geographic NAD83 to the NAD27 state plane using Corpscon but I don’t think that is accurate enough. Then I would still have to convert to the mine’s coordinates that don’t exactly match the state plane system.

There’s no such thing as “own local grid”. Grid references a formal Cartesian coordinate system that is standardized in a region or nation. We need to understand where the mine’s coordinates really came from and reference to as I think was trying to explain. Fact is that the coordinates are either grid, surface or assumed/arbitrary. Grid coordinates that are not traversed correctly become arbitrary and is why so many areas and local surveyors have their own tweaks. Surface coordinates are just grid coordinates that have been scaled to ground and these too become assumed because people don’t know what they are coming from. The only safe way to truly tie into the control without doing massive amounts of research and paperwork is to do a localization. Do you have just a base coordinate or are there other known points to check into? Are they going to use the drone data to integrate with any other systems or design CAD files? If not then it would be best starting with what they have and letting the drone do what it does and let that be the constant. If they are integrating with other systems then the ONLY way you are going to tie in is with a localization.

1 Like

I really appreciate the help here. Some have probably guessed I’m not a professional surveyor. I’m blast designer. I want to use our drone to gather survey data via photogrammetry to import into blast design software as well as CAD software and match it to the mine’s coordinate system.

We’re able to to this already but we have to wait for the mine surveyors to gather points for us whether we use a total station or a drone.

It would be better for everyone if we could gather our own points in the coordinates that the mine uses. The low cost of the Emlids make it feasible if we can convert to the same “grid”.

I don’t know what a localization means. I assume this has been done because there is a lot of equipment there that uses the same coordinates.

If a surveyor can go grab a point with a Trimble rover and it shows the “global” coordinate in latitude/ longitude and the “local grid” coordinate in State Plane, then I should be able to convert the data from the Emlid to match right?

What is a localization?
What information would I need to convert data from a Reach RS+ to the mine’s coordinate system?

When you say gather survey data, what exactly are you trying to import into BDS and CAD? Just the map? Contours? Point cloud?

What points are the surveyors gathering?

I would suggest flying a map and giving it to the surveyors. They will (should) be able to import it into their CAD and figure out the transformation, if it requires a shift or scale factor. Then a template file can be made that can make this an easy three-click adjustment. The drone data can stay on it’s native WGS84 basis, directly exported to the state plane or grid coordinates and then adjusted to their local coordinates. As you said they might have used a localization when setting up their network.

I’m assuming the mine did photogrammetry via airplane in the past? Most modern mines have and it is pretty cool to speak with their managers as we are doing the same thing on a smaller yet highly more accurate scale.

1 Like

If we’re using our laser scanner, we get two points from the mine surveyors, one where the scanner takes it’s measurement and one at the “back site” or “reference”. These are given in the mine’s local coordinates and used in the import into the blast design software. The total station exports an FSC file.

With a drone, we’ve used Drone Deploy in the past, putting out ground control points that are surveyed by the mine. We use those points (given to us in the mine coordinate system) with the Drone Deploy software to get an LAS point cloud.

For now all I want to do (if it’s possible) is get a point with the Emlid and convert it to the mine’s grid coordinates instead of relying on the surveyors.
They must be doing this already but using Trimble.

How would they be doing the conversion?

This is a large mine and does get aerial photogrammetry on a continual basis. I should try to get some information from that service provider if I can. Good idea.

It is very common to use local reference systems to minimize projection deformations. a system is made with a contact meridian in the work area.

1 Like

Good workflow!

To do the conversion would be easy if they are on state plane as it would be a direct transformation and export out of DroneDeploy. If it’s not then you can manually enter the GNSS coordinate of your base from their control to get in place, but that may not account for any rotation. Most mines have large permanent GCP’s, but those are usually spaced across the entire site so if you are flying individual stockpiles or blasting areas then you can’t capture enough of those points to do what you need.

I still suggest letting the drone be on it’s own basis, convert to your regions EPSG for the export out of DroneDeploy and then finalize it in CAD. We have to do this fairly regularly on construction sites even though we are using the site’s control for our GCP’s. It’s a pretty regular occurrence that the engineers use a scale factor which you cannot account for even with GCP’s. A manual shift or rotation is another beast. We actually had a site on 20000,20000 in CAD a couple of months ago that we fixed for the surveyor/engineer after the drone flight. It just depends on what the source files have.

Bottom line is that if you want an easy, efficient and dependable drone program to bring in data that is easy to analyze you probably need to think about updating the sites control using the gear that the program would use. What else are they going to use their control for anyways? We have a 6,000 acre mine/quarry that I just reset the control on for drone purposes in a single day. Once you do that and your Emlid gear has defined the control then you don’t even need GCP’s. You can just fly with PPK.

Do you have the parameters of the custom reference frame your mine uses? It would then be easy to reproject your points with any tool that allows custom coordinate reference systems, like QGIS.

Look for an EPSG:xxxxxxx reference if it’s an unmodified state plane, or try to find out what datum, projection equation, and prime meridian it uses. For example, here’s EPSG:2949, which is a Modified Transverse Mercator with certain meridian, easting and northing values that clamp the reference system to a very local area:

PROJCS[“NAD83(CSRS) / MTM zone 7”,
SPHEROID[“GRS 1980”,6378137,298.257222101,

Your reference system should have parameters similar to these somewhere in your CAD or surveying software.

In my experience there is not going to be a way outside of the CAD package’s environment to do this with utmost confidence. If you have worked extensively in CAD you will know that there are too many variables and eccentricities that can throw the association of GNSS data with CAD Cartesian data off. Luckily for you it sounds like you would only have to do it one or very few times, but after 20 years of CAD and Surveying in Construction and Mining if you truly want to be centimeter accurate you are going to have to either start with the drone or the CAD. Outside solutions are still going to have to be assimilated.

Thanks for all the help! I’ll see what information I can gather about this coordinate system.
I’ll try to update my progress here. Hopefully it will be useful to someone else.

1 Like

In construction with high accuracy demands in large scale we use NTM as a substitute for UTM.
NTM has smaler grid with less scale error then the UTM. The accuracy is better then the overal accuracy of most GNSS recivers. Not sure this is what you were looking for but it helps me do the transition from a CAD to real world locations and vice versa without worrying about the scale error.

I was able to get transformation information and I can now convert from UTM to the local coordinates.


Being in mining are you focused on relativity in your volume management or are you trying to get real surface grade as well?

As far as I know, all volumes are calculated relative to the mine’s coordinate system.

10-4. That’s pretty standard for mines. They initially setup using direct transformation from GNSS WGS84 to their State Plane and Surface elevations, but who knows what happens over time. The last site I flew is about 6000ac, 500ac of which I flew and they had 10ft X’s made out of rubber from the plane photogrammetry the used to due. I setup on one of their control points the looked pretty good, localized and traversed through the two closest local benchmarks. I got a 1.25ft variance so we went with the newer benchmark. The horizontal on the points in the localization were pretty far off too, but were +/- 1ft so for what we were doing it was ok.

I’ve flown a few other mines and they didn’t care about surface elevation or coordinates and the stockpiles were only about 2ac so we didn’t even place GCP’s. Man those are easy except for the fact that there were over 50 piles to take-off.

This topic was automatically closed 100 days after the last reply. New replies are no longer allowed.