Emlid

Community Forum

Basic Workflow for a Newbie?


(Michael Lambert) #24

The bipod adapter, tripod adapter and rover pole are all available in mini versions that adjust from 1-2m. The Rover poler with tripod adapter is used for the base. Rover poles are also available segmented.


(Andrew Yushkevich) #25

Hi @abel.ccaamao,

Yes, it’s better to have a Reach higher as it helps to avoid interference.


#26

Yikes, I may have misunderstood some earlier posts regarding the absolute position accuracy. I need to survey global coordinates (WGS84) of rover points to be 10cm or better accuracy. But now that my units arrived and I am going through the docs, I see for the accuracy I want I need “NTRIP stream from base < 15km” and then the far right column says free/$. I thought I already asked the question about having to subscribe to a service to get the sub DM level accuracy and thought the answer was no just setup the base and let it sit there for 20 min and no PPK post processing necessary - just export the coords when done? But then there’s all this stuff about map shift due to the offset of the base to the actual accurate coordinates. I need the rover to measure the actual world coordinates (WGS84) within 10cm accuracy.

If I’m understanding the docs correctly and I need an NTRIP stream to be < 15km away, how do I find out what’s available in the area I’ll be surveying and how to subscribe to it?

Or am I misunderstanding an “NTRIP stream from base” means NTRIP stream from the base to the rover? I think not as why would it have $ in the far right cost column?

I’m doing my testing in Las Vegas so if I need to subscribe to some service to get absolute position accuracy for my base then that’s the area I’d be doing it in.

And I doubt I’ll have an Internet connection available in the area I’m placing the base unless the NTRIP stream can be transmitted to the base using a cell phone connected to the internet running Reachview? If that’s not how it works, it looks like I am forced to use PPK to correct the absolute coords of the base station by obtaining RINEX logs from a reference station?

Essentially I need to use whatever technique will be the quickest to get me absolute position accuracy in the 5-10cm range.


(Timmyd) #27

Jazee,
In many parts of the country to get NTRIP corrections for RTK it cost you (sometimes alot). In my state (MS) I can get free access for one (and only 1) GPS receiver.

If you learn how to post process your data you don’t need to pay anything for NTRIP service. You simply go out and collect your data (properly and long enough). You can setup your base unit to log and provide corrections to your rover. You can then survey your points with RTK on your Rover and then you will go back and post process the BASE (static) unit to now make your Rover survey absolute accurate and not just relative accurate. If you are new to this, get a big bucket of popcorn and watch these two videos: If you are not new they are full of great information. The guy is an engineer but teaches on a level that all newbies can understand.

Here is the Map to show you where the closes CORS stations are to you. That will let you know how long you need to log in order to get the accuracy you need.

https://www.ngs.noaa.gov/CORS_Map/

This CORS station is 6km from downtown Las Vegas
https://www.ngs.noaa.gov/cgi-cors/corsage.prl?site=NVCA


#28

Thanks. Yes I was aware of CORS. I see you just download the logs from their side then post process it with base station data to correct everything.

So I completed my first field test and made two observations:

At first I was cussing as to why a wi-fi connection was needed as obviously that isn’t always available in the field then I discovered you need to put it in hotspot mode. This wasn’t clear in the documentation I read (but could be because there was more to read I didn’t see.) But then I discovered and issue that was also posted in the forum. You would think that if it can’t connect to WiFi it would default to hotspot. But like another user reported, this is apparently not the case as the only way I could force it to hotspot on power up was to go delete the wifi networks!

Question: Once you have them setup and updated, what is the advantage of connecting over wifi router other than you don’t have to change wifi hotspots to connect to each unit?

So the other thing I need to learn more about is these network choices. The quick start I think said to enable the LORA, which I did. I selected the Average Float when I started a new project. What I’m confused about is in the docs it says Float mode used NTLIB? But the rover screen when I was using it switched from Single to Float after a few minutes. Will it switch to Fixed if I wait even longer and enable that mode on the project setup? Or since I need sub 10cm absolute position accuracy is all of this a moot point and I still need to do PPK using the CORS logs?

I’m a software developer with an engineering degree so this is kind of fun but I was hoping it would be a little more straightforward, or maybe better documented workflows. But everything has a learning curve I guess and I’m not a trained surveyor.

And the last thing that has me puzzled is I set the collection time to 3m and when I hit the collect button, sometimes the progress line disappears before three minutes then you have that striped line and it goes back. When it does this the RMS number are like 0.025m which is 2.5cm which is pretty darn good. But that’s the relative position error I assume, not absolute. So is the data collector prematurely stopping the 3 minute collection time because it already achieved good accuracy without having to wait that long. I think I left the default accuracy for the project for float on 0.5m, not sure if that setting effect the collection progress time line?

Thanks again to all you people throwing me a bone on this. I greatly appreciate it.


(Timmyd) #29

Here is a very good video done by emlid on how to set up your base and Rover for rtk. I will have a complete video hopefully upload it on Saturday that covers every aspect of setting up the reach units, surveying ground control points, and then processing them. This will be done from the perspective of a drone pilot and not a certified surveyor. I hope that after I get this done, that others can do similar videos using different software so that people knew to the reach RS will be able to see the entire process without a single detail left out from start to finish on surveying points.

Emlid RTK video


(Michael Lambert) #30

Unless I am missing something you shouldn’t need a service if you have a base and rover. Letting the base collect the data doesn’t require anything except time. The video that @timmyd supplied is it. Once you have your receivers setup and you have a fixed connection to the base that is where you get your corrections from. Your coordinates are WGS84. I like to take a picture of what is collected by the base so that if I come back to the same site again I can setup on the same point and manually enter those coordinates and just pick up from where I left off.


#31

@chascoadmin

What you are saying seems to contradict the emlid documentation. How can the base find absolute world coordinates with an accuracy better than 10cm without using data from a reference station via NTLIP or COORS RINEX logs using PPK?

See the “Ways to Set The Base” here: https://docs.emlid.com/reachrs/common/tutorials/placing-the-base/

It seems to me what you are describing is the Average Fixed position which only give the base station absolute coordinates accurate to within 2.5m. Yes the relative position of the rover will be accurate to within the 10cm, but the absolute world coordinates of the points surveyed by the rover will be offset by whatever the error is for the base station’s coordinates, up to 2.5m.

The key here is I don’t have a known surveyed point or monument to place the base station on. So to get the absolute position accuracy I need, without having to pay for NTLIP (which I believe requires an Internet connection anyway?) I’ll need to use PPK with the free logs from the nearest COORS station.

The nearest COORS (has a 1 sec rate) will be 10.5km away. So how much time will I need to let the base station sit if I want my accuracy after PPK to end up being say 7cm?


#32

@timmyd

That’s a great basic video and I figured all that out in about a half hour skimming the docs. Too bad there isn’t a video on the data collection part as I’m still wondering why when I set the collection to 3 minutes the progress bar when you hit the collect button goes part of the way, then the hole bar turns into the striped bar and then the progress bar starts over? So like after 30-60 seconds (not 3 minutes) this happens and I note when it does the RMS numbers freeze and are very small like .000x or .00xx I can’t figuire out if some sort of connection error is happening and it is having to restart, or is it just realizing it hit the minimum accuracy before reaching the 3-minute mark? How can I tell when I’ve got like 5cm relative position accuracy to the base when I’m collecting? Just wait until the RMS number goes to .01 or lower? Or do I set the accuracy setting when setting up the project to .05m ? Looks like there’s not a whole lot of documentation (any really) on the data collection workflow.


(Timmyd) #33

I’m now out of the office. I’ll have the video you’re looking for uploaded on Saturday. I’m also hoping to find out good information to use QGIS for post processing as a free option. The program I use, EZserve is very very nice, but it is not free.


#34

Well it seems like the documentation is pretty fragmented all over the Reach site. Have to use Google to find everything. There’s this whole RTK section that wasn’t findable from the main Reach RS doc menu it looks like and found the docs on collection which sort of answered my questions.

https://docs.emlid.com/reach/common/reachview/survey/

What I’m a little miffed at now is that you need to connect to the unit to export the point data? So the data is stored in the receiver and not in the app? I was expecting it to be stored in the app so I could export data without connecting to the receiver. Plus this seems a little wonky having to export using my tablet then transfer to my PC instead. I guess I could connect the PC to the hotspot and go to the receiver web address.


(Timmyd) #35

You can download it to your device (tablet) that is connected to the Reach. I personally just download them to my computer but if you wanted to, you can “download” the log files and survey file to your device and then you can just manage the file on your device as desired.


(Michael Lambert) #36

I see what you’re getting at now. I have never used a service so I don’t quite understand how their definition of position differs from a base station that has been collecting for 20 to 30 minutes. If you don’t have a known survey point then it definitely makes sense. The way it is done in the construction world is a little different because we take whatever GPS we are given by the base and localize it to State Plane coordinates.
As for the transfer of the data I just connect my laptop directly to the receivers hotspot and download it straight to my drive.


#37

@chascoadmin

So Michael, I (like many others based on all the posts I’ve read) am having a hard time getting the rover solution to stay on Fixed instead of float. There are a myriad of settings and dozens and dozens of discussions on this forum of other users having difficulties getting a fixed solution. It’s very hard to troubleshoot without tedious analysis of several logs apparently so it appears the solution to improving the RTK can be unique to the particulars of each user’s application environment.

It seems one can easily find themselves beating their heads against the wall trying to find the “magic combination” of settings to get a fixed instead of float solution!

So I noticed your original reply, you mentioned 2-3 minutes with “even a good float” will get 7cm or better accuracy relative to the base. This is encouraging as I don’t have the time to troubleshoot why my rover is switching between float and fixed all the time. I noted though the default collection time for float is 5 min. So does that mean Reachview defaults to trying to achieve higher accuracy with float than 7cm, therefore the longer default collection time of 5 min compared to 2-3 min?

My other question is, how can you tell when you’ve reached 7cm accuracy when collecting a point? It seems at first, in most cases the RMS numbers while collecting go up!? At least that’s what I’ve observed in a number of cases. Or is your 2-3 minute collection time an estimate based on validating the accuracy by using known points or relative measurements later?


(Michael Lambert) #38

I too had quite some trouble achieving fixes. I have had much better luck since turning my output signal strength down to 16db. The other observation as you mentioned in my area is that we seem to have some GLONASS interference as related to the Reach receivers. I’ve never had an issue with my Topcon, but that is the first constellation I turn off to troubleshoot. My normal congregation is GPS/GLONASS/Galileo/SBAS. There again in my area I do not see much confidence in QZS. Otherwise my setup is typical to their base/rover setup video except for the fact that with those constellations on I have to run the 18.3 kb/s airspeed. You can always have more constellations on for the base, watch the SnR status on the Rover screen and turn off the weakest constellation on the Rover to see if that gets fixed.
As for verifying a float accuracy I use Topcon Magnet Field so I could shoot the point for two minutes, back for a second and then stake that point out.This is now accomplishable in the new version of Reachview! If my stakeout of the point was not acceptable then I would reshoot the point and stake again. This could be fruitless, but in my experience I would consistently get within 0.2’ which is fine for my drone work. At least I wasn’t washed-out if I couldn’t get a fix. Once I have set the GCPs I always check back in to my first point. Once that network is set then that project is good as long as I got my points in areas that wouldn’t get disturbed. My drone surveys are then repeatable.
My last hiccup that I am working through with Emlid is that now I am achieving fixes regularly and the Topcon system won’t let me shoot. Something in the communication between the Reach and Topcon software, but I can see the I have a fix and that the coordinates are not moving so I just record it from what I see on screen until we figure it out. I stick with this setup because my end goal is for each of our 30 crews to have their own base/rover and the ability to use the same software which allows us to view our CAD file linework for the project. This would increase our survey labor efficiencies by a huge percentage because it would remove the need to drive for 30-60 minutes to lay out 10 points or replace something that has gotten knocked out.


Can't Get a Fix - Any Published Best Practice Troubleshooting Steps?
(Andrew Yushkevich) #39

Hi @jazee

The answer is that it’s much easy to work with if it’s in one network with your PC. Also, you’ll have permanent Internet access. However, it’s your choice.

Yes, Reach will automatically switch on another mode but the averaging of the point will start from the beginning.

You can stop it by yourself if you need. Reach will average all positions during these 3 minutes even if it’ll get RMS earlier.


#40

So now that I have a better understanding (over the hump on the learning curve I hope), I realize the resetting of the progress bar on the collection was just due to the rover losing the fix and switching to float.

As I mentioned, on float, the default time is 5 min, and on fix it is 40 seconds. If I’m going for 7cm accuracy, could I therefore on fix get away with like 20-30 seconds probably, or even less!?


(Michael Lambert) #41

I think once you become comfortable with understanding the readings that is totally fine in the right conditions. My experience with the Topcon is that it took about 100 individual shots in 1 minute. The one thing that is nice about coupling with Topcon is that it has crude onboard post-processing so it probably provides a little better end result than what we currently get with Reachview.


#42

I’m desperate to find the step by step to post process a survey (correct base station coords with using COORS logs), then correct the coords of the surveyed points and output a CSV with the corrected points. Pulling my hair out trying to find docs. The Emlid post processing doc for RTKLib is okay but it only briefly mentions correcting the base and doesn’t cover it as a two-step process to eventually get the corrected rover points. Found a forum example of the process to correct the base but it was for then going back out to do the survey and manually entering the base coords.

I don’t want this thread to get to big and go off on tangents so I’ve started a new thread about it here:


(system) closed #44

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