High Accuracy RTK Positions with the LEA-6 using RTCM via NTRIP

I was just thinking...

Has anyone looking into real-time correcting the LEA-6 (or other DIY Drones GPS) using RTCM via Internet Protocol (NTRIP).  I'm assuming you would just need to have a cellular modem (or phone) on the copter connected to the internet, receiving the RTCM data and passing it along to the GPS unit.  I checked the U-Blox website and it says the LEA-6 has an interface for the RTCM protocol (http://www.u-blox.com/images/downloads/Product_Docs/LEA-6N_ProductS...).  Many state's now offer a real-time GNSS network via NTRIP, some are free subscriptions while others are paid subscriptions.

I'm not a programmer so I'm unsure what it would require on that end to make it work.

Here is some example code I found for .NET (http://www.sharpgis.net/post/2005/01/21/Differential-GPS-using-RTCM...)

Any thoughts?


Views: 19222

Reply to This

Replies to This Discussion

What I would have given to know about this stuff 3 years ago... 

Definitely looks feasible. On my list of things to test out. First I need to buy a plane... 

I'd be willing to write the software if someone else furnishes the bird. 

Why use this?  Same data comes from WAAS.  Why not make sure that's working first?

I don't think that's even currently enabled in the MT GPS on the APM.


WAAS is a Wide-Area Differential Global Positioning System (DGPS) or SBAS (Satellite Based Augmentation System).  You typically will see 1 m laterally and 1.5 m vertically accurate positions, best case.

What I'm talking about is a DGPS that uses a network of fixed, ground-based reference stations.  The reference stations that are broadcasting corrections have been set to mm accurate positions.  You can get 1 cm laterally and 2 cm vertically accurate positions, best case.

I have used the service in my area and I know that it supports both single-frequency and dual-frequency receivers.  I have only connected a dual-frequency receiver to the network and would like to see if a single-frequency receiver (like the LEA-6) would work and what kind of accuracies you could get.  However, I don't have the software to make the connection.

What you're talking about is essentially the same system as far as I understand.  The WAAS gets it's data from the same ground based reference stations.

The ionosphere is the main source of GPS error.  Ground based reference stations track this error and send the correction data to the satellites.  The satellites then send this data out in their signals.  That's called WAAS.

The only difference in the system you're describing is that the data is arriving from the internet rather than over the GPS satellite signals.  That's why I'm not understanding what the point is.  How is using the internet better than using the WAAS satellite signals?

There has to be some reason or people wouldn't be doing it.  But I think the main reason is that people have their own reference stations that are closer to where they're operating rather than using the public WAAS reference stations, which might be further away from their location, update slower, or have some other limitations.

So I can see why you might want to do something like this if you want your own reference station.  Why not just use a GPS at your GCS?  You could correct once at a certain reference point, then move your station to the operating location and then use drift correction from there.  You wouldn't need the internet to be involved to do something like that.

The WAAS ground-based network is composed of about 20 or so reference stations in the CONUS.  The networks I'm talking about are composed of a grid of reference stations spaced every 50 km, or so, around each state (that has one).  The one I've used has around 80 reference stations spread throughout the state and also shares with the stations for the bordering state's networks.  Using a dense grid like this, will lower the ionosphere error effects dramatically.

You could use a GPS "base station" at your GCS, but you would need it precisely setup on a known point with a highly-accurate horizontal and vertical position already established (if you want it to be accurate).  Being that there are already so many Continuously Operating Reference Stations (CORS) Networks around the CONUS, why not tap into those.  Then you don't need to have a second GPS and a known point to setup over.

Thanks for the explanation.  I wonder why these stations aren't incorporated into the WAAS network?  Maybe there just isn't enough bandwidth on the sat signals to do more than a few dozen points.

Sounds like you have some good ideas, I just don't usually have internet access where I fly.  I'd be more interested in a personal reference type setup.  I don't think it should be that hard to find good reference points, especially if you had mapped your flying area and referenced it to google earth.  I would think it would be pretty easy to "zero" your gps at a known point, move to the actual area you want to operate from, then correct for any further drift.

With a $30 GPS unit for your GCS you could save the $30/month data plan for the cell phone internet service.  Just an idea.

Do you have any links to info on these reference station networks?  I'm curious to see if there's any in my area. 


The WAAS network is setup specifically for the FAA and they want to make sure it is 99.999% reliable so when you're landing an airplane, you don't lose your corrections.  They are only guaranteed accurate to 7.6 m accuracy (horizontal and vertical) at the 95% confidence level, which is just fine to land a plane.

The networks I'm talking about are setup by either Federal, State, City, or private entities and are not as reliable (they are still functional 99% of the time).  But, they are much more dense and much more accurate when you receive corrections and are typically guaranteed to cm accuracy (horizontal and vertical) at the 95% confidence level.

Google Earth is not very accurate.  Maybe only a few meters at best in most places.  I would like to see a 10 cm or better accurate GNSS solution for the copters.  Which is why I was wondering about the RTCM via NTRIP corrections.  Most people have a data package with their phone and could use it for the internet connection.

I would also like to see the DGPS solution using the GCS as you state (especially in areas where cellular reception is limited).  I have used that type of setup with dual-frequency receivers and it works great, cm accurate as well.  The problem is always finding a highly-accurate (a few cm's or better) point to use as a reference for the receivers.  If you have a dual-frequency (pretty expensive) receiver, or one that can store a RINEX file, than you can collect 15-minutes (or more) of data and post-process the coordinates to within a few cm's using the NGS OPUS-RS online program (http://www.ngs.noaa.gov/OPUS/).  Then you can have a highly-accurate point anywhere in the CONUS.  Otherwise, highly-accurate points are hard to come by in most locations.

Here is a link to the CORS stations that are registered (or Blue Booked) with the NGS (National Geodetic Survey) http://geodesy.noaa.gov/CORS/GoogleMap/CORS.shtml.  As you can see, it is a pretty dense network.  Typically, if there are a lot of CORS stations in a state, then they have some kind of RTN (real-time network) setup where you can receive corrections (either via NTRIP or radio).  Sometimes they are owned by public or private entities and sometimes they have a fee associated with them.

Sounds like a great idea!  The more accurate you can get your GPS the more things you can use it for.

I wonder if you could use a cheap SDR to pickup radio correction signals?  I've been looking at getting one to pick up ADSB signals.


I can't seem to find any info on radio transmission though, so maybe the internet is the only source for this data.

Nearest site is about 8km away from me.  I wonder how much help that would provide?

I'm not really concerned with cm accuracy.  My main concern is that the google earth imagery location and my GPS location matches up, and I'd be very happy with 1m resolution.  1 foot accuracy would enable just about every use I can think of.

Yes, GPS accuracy is important.

I'm not sure about what all can be used to pick up the signals.  I do know that they typically broadcast in the 450-470 MHz range.  All you need is the exact frequency they are broadcasting and then you should be able to receive them, with the right hardware/software setup.

8km isn't too far.  Typically (if they use radios) the signal will carry for about 10 km, depending on the terrain and the wattage of the broadcasting radio.

The issue you might have with google earth imagery is that once you get an accurate GPS, you'll realize that google earth imagery isn't always that accurate and you can't match it.

Do you know anyone who is working on getting the GPS at a GCS working with the GPS in the copter?  I'd be glad to help test it out or help in any other way I can.

I'm not aware that anyone is working on a GPS correction scheme.  The "follow me" feature is coordinating two GPS locations though, I think.

I would love to get real time <10cm accuracy.

Another option would be using some post processing.  Is there a way to log the raw GPS observations in the APM2 with LEA-6, and post process (RTKLIB) with base station data.  Then update the log file with the corrected information? 

I would be happy with either process.

I agree, the ability to post-process would be a great start.  Real-time corrections, either though the internet or via radio link to another receiver, would be the best.

Now if we can get someone who knows how to write the code and make it all happen.

Does DIY Drones have a GPS software developer in their crew?

Reply to Discussion



Season Two of the Trust Time Trial (T3) Contest 
A list of all T3 contests is here. The current round, the Vertical Horizontal one, is here

© 2020   Created by Chris Anderson.   Powered by

Badges  |  Report an Issue  |  Terms of Service