Something I want my RTK GPS to do

I believe Topcon TopSurv & Magnet will do this, though I have never had the courage to try it. I believe that when measure to the last known control point, you save it using the same point name it already has. TopSurv / Magnet will then ask if you want to overwrite, or average the point or Correct the Base. I believe that last option will do what you want.

Nate, why are you worrying about stuff like this??? It's RTK, so as long as you are generating numbers, you've got you some solid survey results that probably can be figured out either in the office or in preparation for some future deposition. :>

> I believe Topcon TopSurv & Magnet will do this, though I have never had the courage to try it. I believe that when measure to the last known control point, you save it using the same point name it already has. TopSurv / Magnet will then ask if you want to overwrite, or average the point or Correct the Base. I believe that last option will do what you want.

Byanez you are correct. Works best if you are using a state plane projection, but can not use a vrs. So no rotation need. Just transformation. I have used that routine many times. Setup your base on an unknown point. Then measure a known point in Survey Topo. You will get prompt that point already exist. Showing you delta n,e,z Choose last option "correct base". Now you are tied in.

First off, you shouldn't use RTK for control.

I would static in the new base from the old base.

That's it, now I can occupy the new base in the same coordinate system.

In a post about setting up a small local network to localize to, and essentially this is the same thing. As I understand it, what you want to do is fire up your unit, occupy the coords you've already processed for the other job, and get to work with good control and no worries. Localization is what it is called around here, not sure if that term is nationwide but the collector we use, Leica CS15, this can be done with a small process that if you knew the procedure would probably only take a short amount of time.

Here is the link: http://spatialtechnologies.ca/wp-content/uploads/2012/04/determine-coordinate-system-onestep-multiple-point.pdf

Not sure what unit you are using but I would imagine this option is available in all modern RTK collectors.

Lee,
does the rover communicate back to the base to translate the actual base position or does the rover apply the translation to the incoming offset corrections?

> Lee,
> does the rover communicate back to the base to translate the actual base position or does the rover apply the translation to the incoming offset corrections?

Shawn,

No communication from rover to base. The controller calcs the dN, dE, dZ and will recompute your raw data using the corrected base coordinate. So only the data in your controller is adjusted. In raw database you will the original base coordinate, and the corrected base. You can use Tools to make an adjustment in office if you find it is needed.

Lee Green

Thanks for the explanation!

Are you not working in SPC or some other well defined coordinate system?

If you were, this would called a "one-point localization" in TDS, I believe, and be an available when you start the job.

I'm working in my own low distortion projection. This is useful for small areas. Say, 25 miles, wide, and and some length.

I take time, to get nice long shots, in clear areas. So far, it has worked well. I know that every time I move the base, I get a bit of error. But the closer the BASE is to the site, the better. So, a NEW base set up has advantages.

Post processed is better. I have networks out there, from the LOCUS days, They are LDP, on State Plane Brngs.

But, sometimes LOCALIZE is not the best answer, because I want to CONTINUE the projection.

N

I work in SPC, but the correct base will work in a localization also.

You are missing the concept. Let me try to explain what Nate is doing.

Nate has a boundary job from 2010 in SPC. In 2015, today Nate now has new job just 1/2 mile down the road from the 2010 job. He has a map in hand from old with SPC on it. He setup his base today on an unknown point, in open sky near the new job. Nate does NOT have cell coverage. So he has to use an autonomous position at the base. It's about ±15ft from SPC, sometimes better. This get's him in ball park to start reconn. nate heads down the road to the old 2010 job. Nate finds a good iron that is in open sky, good for GPS. Occupies that point, it is point 100 with known coordinates in controller already. Nate sets the ROVER on point 100, in Magnet go to Survey > Topo enter point 100, hit measure. Being that point 100 is already in Magnet, the software prompts the user that point already exist.

nate is given 4 options to choose from:
1) Rename point
2) Over write Point
3) Use point as Weight Average
4) Correct BASE

Nate chose option 4 - Correct Base. The software will adjust the base coordinate, so the dN, dE, dZ are now zero at point 100. Then Magnet will recompute the raw from this corrected base coordinate. Nate is now tied into to the same coordinate system of old 2010 job using RTK. Nate did NOT move his base. No need to localize. I would suggest Nate continues to search and verify a few more points.

Lee Green

Lee,

You are right on.

That's exactly what I'm up to!

N

If your data collector doesn't do it for you, there is a pretty simple process to accomplish what you want.

Set on new point 201, call it something else, say X201, locate 1, call it X1, inverse from X1 to 1, then calculate point base point 201 from X201 using the same inverse numbers inversed from X1 to 1.

Of course once you have the data there are a number of ways to shift X201 to 201 using the same shift between X1 and 1.

Restart the base using 201 as the base point.

It takes a bit of time (not all that much really), but you now have the point you want, in the projection you want, and you are broadcasting the correct lats and longs while you do your survey, you don't need to worry about how the data collector is shifting things around.

Once you have an LDP established one thing you don't want to do it tweak it doing a calibration, BAD idea.

>> Nate has a boundary job from 2010 in SPC ....

That part of your narrative is wrong. If he were working in SPC, then I think it would be much easier, but as he said above, he's not, he's working on a custom grid.

I've never tried it on a custom grid, but I bet what you described with Magnet, and what I've done in the past in TDS, are both basically the same thing and yield the same result with the same amount of effort.

In TDS, the procedure you described is called a "one-point localization", where you set your base up anywhere, "side-shoot" a known control point, then it automatically adjusts your base point to match that coordinate.

Only thing is, I'm pretty sure with TDS you have to start a new job in order to get that option. So, what I would do (back when I used to do things like this) was start a new job, import my existing Ascii file, then start the GPS survey and do the one point localization ...

The Magnet version sounds pretty slick, though, since you don't have to create new job each time you want to do that ... so I guess the advice Nate gets depends on which DC software he's using ... I think it's Carlson.

Now just one problem. Nate has a Javad system, and it will not work with Topcon's Magnet software.

If we could just get Javad's hardware/software to merge with Topcon's Magnet software.

Topcon/Magnet needs from Javad
UHF radio scanner
Gyro no rod bubble feature
Camera feature
6 RTK engine feature

Javad needs from Topcon
Magnet software intuitive, user friendly ease of use
Magnet's mapping features
ToughPad Windows controller

Topcon and Javad need from Trimble
R10 Scanner function

Would be nice to copy/paste the features we like from each vendor into one system.

Localizations perform one of two distinct operations. (Anyone that knows better, feel free to correct me).

Operation One: Create Custom Projection.
In this operation, the user creates a projection. This may be the form of a single point projection, such as setting up the base on a "here" (autonomous) location and stating the grid coordinates of the base are 5000, 5000. The software then creates a projection with the geodetic (lat/long) coordinates of this point as the origin and the Grid North and Grid East as (for instance) 5000,5000, with a Geodetic North bearing relation at this origin. I don't know what projection type Topcon employs (their documentation is buried in their website), but Trimble uses Transverse Mercator for these projections, while Javad uses Oblique Stereographic. Carlson's documentation doesn't appear to specify what type of projection it creates in this operation.

Operation Two: Determine Transformation Parameters.
In this operation, the user is working in a known projection. It could be State Plane or any other defined projection recognized by the data collector. If the projection is known, a single point will suffice (although it's not necessarily recommended to rely on a single point without some redundant check). In this, the projection type is known, a Helmert transformation is applied to determine the translation (if there is only a single point) or the translation, rotation and scale (if there is more than one point). Typically, in a known projection, you don't want to distort rotation and scale as these are "known" parts of the projection, so rather than allow the transformation to apply these parts of the Helmert transformation, you are often better to force these to the projection values and let the residuals of the coordinates increase as a result. If the points you're localizing on are precise and the survey points are precise, the residuals shouldn't be terribly large.

There's a blending of these two operations when dealing with localizing to multiple points in an unknown system, such as retracing a total station survey. Operation one creates a projection at the centroid of the localization points (simple average of all Northings and of all Eastings). Then the transformation parameters are determined (rotation and scale) to best fit the GPS derived coordinates in that localization to the unknown grid. In this case, you'll want to let the rotation float, most likely, unless the total station survey's precise relationship to geodetic north is known. Scale should be carefully considered. Most likely, you will not want to use the scale factor determined, because the GPS survey is generally more accurate than the total station survey distances (particularly on longer baselines) and the scale will distort the "better" GPS measurements. But this isn't always the case. Professional discretion is very important when it comes to localizations.

Operation Two, is basically what you're after, Nate. And, so far as I know, it's in all RTK collector software. You have a base broadcasting corrections from an autonomous position in a known projection (local or otherwise, both are a "Grid"). You tie in a single point and localize, which provides you with a translation in 3D (xyz or neu). This translation is applied to every point collected with this localization in effect. This is where localizations get dangerous, as this localization only applies to this base point, broadcasting this particular autonomously derived position. Move the base to another point (determined by autonomous OR determined by observation using the localization) and you are applying an INVALID translation. Bad stuff.

Hope this helps.

I want mine to go out in the field and collect all measurement then return to the office or home then load and analyze the data on the PC for me, and then report to me, then set it self back on it's charger.

I want a function built into the controller software that initiates the sound of an air raid siren when it begins using an autonomous position for the base instead of the corrected position that I entered. Maybe the sound of screeching brakes and a multi vehicular pile up for the sound of a bogus fix? Or maybe a police siren when I forget to check in to control after taking multiple shots?

B-)

If it was me doing this, I would go back to the base and restart it from the new calculated coordinates before taking any shots on the new survey you are starting.

I can see a lot of benefit from being able to do it this way.

James