RTK Correction Data from RS2+ to Drone

I use an Emlid RS2+ base on a known point to connect to my M3E and M350.

I first connect the remote control of the drone to the RS2+ hotspot. My Emlid correction input is turned off because my base location was entered manually. My base output is “local NTRIP.” On the DJI Pilot 2 app, I enter the credentials of the base.

Do I have to keep my remote control close to the base to maintain the connection to the base hotspot to ensure that RTK data is getting received by the drone? Or does the drone receive RTK data from the base? I have a long (1 mile) and relatively narrow area that I want to break up into two smaller missions (with overlap). I’d like to fly the first flight next to the base, but the second flight near the middle of my area of interest so I can maintain line of sight with the drone. Will the RS2+ still be sending RTK correction data to the drone when my RC and I are 1/2 mile from the base? I’ve seen conflicting information about this and am looking for clarification.

if you are using the RS2+'s hotspot, then you must keep the Drone RC close to the RS2+.

In the case where you want to do two overlapping sectors/flights, then you’d need to set up your base inside that overlapping area and use it as a location to launch both flights from.

If you can’t keep the RC close to the bases’ hotspot, then you’ll need to look into using Mobile/Cellular access on both the RS2+ and the RC unit, and relaying the NTRIP from the RS2+ through the mobile network to the RC unit where you are flying. You will also need to utilise the Emlid Caster service to help in the relaying of the NTRIP stream to the RC. ie RS2±>Mobile SIM->Emlid Caster In->Emlid Caster Out->Phone or Router with Mobile SIM->Phone Hotspot or Router WIFI->RC Controller.

In all cases, the drone’s RC first receives the NTRIP stream from the RS unit, and then the RC unit relays the NTRIP stream to the drone using DJI Ocusync system. The RS2+ never communicates directly with the drone.

Hope that helps.

3 Likes

Thanks for clearing that up Nick. I can’t use mobile/cellular access for this project because there is no cell coverage. It’s quite remote. Is there a way I can amplify or broadcast the NTRIP data from the RS2+ to the RC without cell or wifi? I should be fine setting up my base inside the overlapping area of both flights. Which leads to my next question:

I have a known point that a surveyor set up at this site, at one the far ends of my entire flight area. If I set up my base (RS2+) on this point and enter the coordinates manually output that correction via lora and then record my new base location with the rover (RS2+), will that be good enough for my base location for the RTK flight (and to record raw GNSS data)? Or would it be preferable to record this new base location with the receiver and post process in OPUS?

Hi Jason,

I echo @ElectroNick that the base (RS2+) connect to the RC and RC relay to the drone via telemetry.

To extend the distance between the Reach RS2+ and the remote control, you can use the Local NTRIP. Instead of using the Reach hotspot, you can also use the hotspot from your mobile device. This hotspot doesn’t need to have an internet connection. Will this work for your workflow?

Thank you Merryna. I have been using the local NTRIP to connect the RS2+ and the RC, but the RC doesn’t get RTK corrections until I connect the RC to the RS2+ via the RS2+ hotspot. I’ve tried this with RCs from both the M3E and the M350.

There is no other way to connect without an internetconnection than you say here. Well, there is, but they are very cumbersome and hardware-requiring (LoRa etc).

Sure, you can do a another point somewhere, but remember that each you create a new point from another point, you have to add your setup and ppm error budget to the new position.
As such, it seems to me you’ll be better off with using PPK for this particular job?

3 Likes

I’m not sure what the best practice here would be for you.

There are several ways to do the job, and it depends on what hardware you have available to you, and how much time you want to spend preparing and running the job.

For a single static RS2+ base over your surveyor’s known point outputting an NTRIP stream to your drone’s controller, you’re going to need some way to relay the NTRIP stream to the further reaches of your area. If you’ve no cell coverage then you’re going to have to try using one (or more!) wifi-router/repeaters. GL.iNet make several different versions which you can power from usb batteries, or from an internal battery. The Routers can use cell sim cards to access 3g/4g, but can also work to create a stand-alone WiFi network without a sim. The repeater range varies from model to model. The Router I use manages about a 50-100m radius (depending on what’s blocking the signal at each end). You’d need to set the routers to create their own WiFi bridging network, then overlap each router’s area until you’ve reached your other take-off/flying zones. In this scenario, you’d first create the Router WiFi network and get everything up & connected, then start up the RS2+ and connect it to that router WiFi network. Then you’d start up your RC and have it connect to the RS2+ via the Router WiFi.

RS2+ - GL.Inet WiFi - GL.Inet WiFi - Gl…iNet Wifi - RC Controller < drone.

FWIW, I nearly always use a GL.Inet portable router to act as a central hotspot now. I’ve found it to be far less hassle to get everything connected than using the RS3’s hotspot, or my phone’s hotspot. If I don’t use the GL.iNet I find I can end up spending 15-20 mins trying to get everything talking to each other as the RC can get grumpy about the order things join the WiFi network. With the GL.iNet I simply make sure it’s the first thing turned on, then everything else simply finds it and connects when they’re turned on later. It makes it a lot easier to keep the drone in sight during a flight too as it frees you up to walk further away from the RS base without having to worry too much about losing the WiFi and having the NTRIP stream drop out.

A different way would be to use two RS units in base-rover (using LoRa or other radio) configuration to lay out several ‘known’ points across your survey area. Some of those points would be utilised for GCP’s and CP’s, reserving one central point to act as your new local base. Then when you come to fly the drone, you setup your RS2+ as a Base with local NTRIP on that measured central point and fly all missions from that central base. This way both missions are working from a single common base, and both lots of data will be locally accurate to that point. You can then PPK that central base point if you wanted either to check its accuracy, or to recalculate it if you wanted. If you marked and photographed your surveyor’s original known point in the drone flight, you could also use that as a GCP/CP to help validate your in-flight data.

Yet another way is simply to ignore the surveyors point for now, set up your RS2+ base centrally, and fly the missions from there. Then PPK your RS2+'s position, and use that to PPK your drone data. You could also occupy the known point with your RS2+ and take some additional static readings from it, PPK’ing them to obtain a fixed point and then seeing what difference you get in readings from the surveyor’s known point. In some drone PPK software, you can then offset the drone readings to bring them in line with the known point.

A final way is to set out several GCP/CP’s around the survey area and take static readings for each those points - you’d then PPK each of these points later. Then fly your missions moving your local RS2+ NTRIP base between each mission. you can then ppk each separate base-Mission dataset individually, and then use the PPK’d GCP/CP’s in your modelling software to conform the model to the GCP locations.

I’m sure someone else will chime in with some extra do’s and don’ts for each of those scenarios - I’m bound to have missed something obvious!

Hope that doesn’t confuse you! It’s certainly a challenge to work out the best method for doing larger areas without access to the mobile networks.

(FWIW, I’ll be using a mixture of the base-rover lora gcps and static averaged gcps to do a survey for a very large heritage mine site. I’ve not quite figured the best method yet and probably will only be able to figure it out once I’ve walked the job first to see what areas get mobile or lora coverage, and what areas get nothing (lots of ghylls and hills). It’s probably going to take me at least four days, one just to walk it all first, then another day to set out and measure gcps and cps, then a couple of days to fly the sectors. I’ve already spent at least a day looking at the maps and trying various mission configs in FlightHub2 as that will affect where I put the GCP’s.)

2 Likes

Thank you @ElectroNick for explaining this to those who are newbies. A lot of people don’t realize that it takes time to explain any kind of methodology that associates with the surveying profession.

Many newbies expect all the hardware and software to do the work for them, but they also need to learn the basics to proceed further in this profession.

You can’t always rely on other people’s knowledge without the basics.

3 Likes

thanks Bryan - I’m still a newbie myself, only having started to try and understand how it all works together less than a year ago. It’s been a steep learning curve, but I’m getting there. But then I’ll never stop learning until I stop breathing permanently!

2 Likes

Me, too ! I’m 68 and new survey technology keeps me going as I enjoy learning new things. Been in this profession since my dad started our business in 1972, we’ve come a long way since transit and tape methodology. I feel we’re at the leading edge with equipment and software but there’s more out there to learn for sure.

For example, I’m studying for my FCC technician class ham radio license now. It’s surely different from land surveying but I got to put to use my new Xiegu X6200 transceiver I got for Christmas . I asked my wife why she got it for me and she said I need to start learning something other than land surveying .

2 Likes

Yes. And that is essentially what @ElectroNick is referring to in the following part of his reply.

This is what I would probably do if it were me. Set up on the surveyed point and NTRIP over LoRa to a new point central to the project that will be used as your base for both flight areas. After you get all done, and wait a couple of days, you could check your new base point against OPUS and see how close your new base point is to what NGS spits out. (Flow 360 makes this really easy with the Inverse tool). If it is within your error budget, you are done and ready to process your imagery. If it is not, then you have the option to post process your images locations with the corrected info from OPUS before processing the imagery.

2 Likes

Thanks Nick. That’s a wealth of information. That sounds like an epic job you have coming up. Good luck.

Thank you Dave!