Land Survey Possible? post process?

Hi, Im a land surveyor, and i have experience (10 years) using gnss equipment for land surveying, my land survey equipment have a millimeter accuracy in RTK, but they cost around $20,000 USD, and the drones on the market for photogrametric collection cost the triple or more.

I think there is a huge market for this product in this area (land survey). but it needs to emerge with somebody that know land survey and somebody that knows how to program this RTK and GNSS data, because i get lost with all the interactions that this product can have or the programming code options.

i would like to know if this raw data can be post-processed to get millimetric measurements.



what you are talking about surely is an interesting market opportunity. Though we are usually talking about centimeter accuracy rather than millimeter with this type of “hobby” equipment (I mean Emlid products for example) and the overall “hobby industry”. The level of accuracy probably goes back to more factors like following:

  1. RTK hardware quality (GNSS modules and antennas used)
  2. RTK software quality
  3. surveying software quality
  4. surveying know-how
  5. external conditions like weather etc.

I think that with surveying there may usually be multiple GNSS (GPS) bands/frequencies used (not only L1, but also L2 etc.) where you get much better accuracy but for multiple bands the cost of receivers and antennas then skyrockets probably (from $100s to $1000s). The hobby market mostly is single frequency equipment (like GPS L1) and the receivers are mass produced so the cost can by driven down. The question is whether one could create a reasonable surveying equipment with a single frequency hardware.

Regarding post processing to get millimeter precision - surely the output of hobby equipment can be postprocessed but I am not sure about the best precision. Why is a millimeter precision needed for land surveying? Centimeters are not enough?

Data can be post-processed, getting logs is actually very straightforward using Reach View. We are preparing a tutorial how to post-process raw data logs. No code writing is required for that.


Thank you for answering my questions kozuch, now i think that good surveying equipment can be created with this hardware and without getting the skyrockets prices.

the type of jobs to be made with the drone can be classified in 3 different types:

  1. The hardest one to make, with this equipment is placing new marks (millimetric accuracy required) or stakes of a property, this requires to load the marks on your GNSS equipment. this marks are taken from the property tile, some times the tile have the utm coordinates in the other case we use some CAD software to get them.

  2. Just collecting data on the field where ever is needed, like property marks, already settled down or just different places to get slope contours of the property. it can be post processed later and no problem with this one.

  3. And the easiest one the photogrammetric surveying where the hardest part its processing the pictures with pc software and get measurement or 3D contour lines. here the millimeters are not needed.

the first one is the challenging to get, and it wont work in some areas, with vegetation, but there are many that will work on, like desert, beach or new construction areas where the machinery cleaned all vegetation out.

the way i think it can be possibly made with this centimetric accuracy is to land the drone in the mark and then wait a while to collect some raw data, get the data via GSM in your PC and process it to make it millimetric and make the corrections and get the distance an angle needed from the center of the drone, since the real mark should be a few centimeters around, something like an arm to control a small hose around to the desired spot and make a mark.

for the mark i think that something like polyurethane foam should be used, because the light weight, and it will stick to most surfaces, in the beach sand could be a problem for this method.

i dont know much of design but this drone will need to have 1 foot or more, of ground clearance to have a good camera view and for making the marks with some movable arm around the ground for the sprayer.

number 2 and 3 looks that they can be done without problem. where ever you can land it you can get raw data for that spot, and the photogrammetric you can configure the drone to take the picture at the desired coordinates and altitude and make some reference mark in the ground and you are ready to go. this photogrametric tool looks very interesting, since you can cover big areas or hard to get, at very low cost, compared to the cost that survey companies will charge for this.

something else that can be used in this area, is the LIDAR sensor to make easier calculations or measurement, but again I don´t know much of the combinations that can be made with this feature and the programming it

4 posts were split to a new topic: Not able to convert Reach log to Rinex

Hi! I do work as a land surveyor, and I can say L1 material is also used by land surveyors for its lower price. L1-only hardware suffer from these two limitations :

  • longer initialization time (comparing to L2)
  • good accuracy but only with short baselines, whereas L1-L2 receivers can achieve a better accuracy with high baselines

I’m very much interested by your product, and I could buy some for my company and do some accuracy experimentations.

1 Like

Hi Igor, Do you have the tutorial on post processing available?

Hi Igor! The height of the antenna (base and rover) can be inserted in settings? For land surveys this is an important setting.

1 Like

What point on the GPS antenna do you measure to? The ground plane or some other point on the antenna? Other references talk about measuring to the “phase center of the antenna.” Any idea where that is for the supplied antenna?

I have also been wondering if the supplied GPS antenna needs to be “calibrated” (also mentioned in other references), which I assume means finding the phase center of the antenna.


1 Like

I don’t know if you saw this post, if it helps:

1 Like

The antenna height is not a big deal, this height is only the distance that the software is going to automatically subtract from the Z coordinates (elevation) to get the ground level coordinate. you can also do this manually, you have to measure from the ground or stake point you are going to measure, to the antenna phase center and substract it in the Z value of the coordinates that the antenna measured. the purpose of this is to get the ground Z coordinate and not the tripod Z coordinate.

the antenna phase center is the point on the antenna that is going to use as a Z coordinate.

You only gonna need this when you want to measure the Z coordinates or heights between points. in the other hand when you only need to measure distances in X and Y, you don need to worry about this, just ignore the Z coordinates (elevation) and use the X and Y (lat-long).

In case you need to measure Z coordinates, i recomend to leave the default height (zero or whatever) in the rover’s antenna and use the same pole height to measure all the points you need and there you go, all your rover’s Z coordinates (dont use your base antenna Z) collected will have the real difference in Z between them, i do this when i use my GPS survey equipment and everything is OK.

1 Like

Hi, can anyone share the steps involved to post process the Reach Logs using RTKLIB. Would like to post process our ground control mats for our aerial UAV surveys.

Kyle, we have plans to do a tutorial on post processing, but the weather is terrible to go out and collect good data. If someone could provide us with some logs from Reach we will be happy to use them.

No problems, I can help. Do I simply place my 2 units out in a paddock for a couple of hours then send you the 2 logs?

This will be great! Actually I was thinking about a dynamic test. We have antennas set up on the roof, so collecting static data is not an issue.

If you could install the base stationary and walk a little with the second unit in an open area that would be awesome. I guess you are in AU? Then it would be better if you record GPS+Beidou data.

ok, will do, thanks.

1 Like

Igor I am in Brazil. Should I record GPS + GLONASS data?

Yes, GPS+Glonass would be beneficial for Brazil.

Tks Igor! What would be the best update rate to record GPS + GLONASS data?

5Hz would be best for a dynamic test.