RTK with Android (RTKLIB + correction data)

[israel kariti]

I am very curious about your opinion about rtk with android.
How accurate can we get with RTK based on Android?

Assuming we have turned off duty cycling, having access to raw gnss measurements, and also accessing some correction data from a paid/free base station, and applying it all into RTKLIB.
Do you believe that it is possible to get good fix with, let's say, pixel 4, assuming that i have access to correction data from nearby stations?  

Can we get centimeter accuracy? or something close? and if so - would it be possible for open sky or even urban canyons? Tnx

[gpsfan]

 But last I checked RTKDroid could not connect to the internal receiver ? I had managed to connect to an external receiver (old uBlox BT receiver and uBlox 8 USB receiver) but it was not very convenient. Is there a new version that's compatible ?

[israel kariti]

My question is actually 2 part question
1) Do you believe that it is possible to create myself an RTKDroid alternative that utilizes the internal receiver raw gnss measurements?
2) How accurate do you think this adventure can be? (Assuming duty cycling is off + access to base station that is very very close to me e.g. 1 Km)

[Sean Barbeau]

This paper looks like a good start on the subject (although I haven't had a chance to read it yet):

https://www.sciencedirect.com/science/article/pii/S0263224121001858

I've been trying to track relevant academic papers here:

https://github.com/barbeau/awesome-gnss/blob/main/README.md#academic-peer-reviewed-publications

Please let me know if you find others you think I should add.

Sean

--
You received this message because you are subscribed to the Google Groups "GPSTest" group.
To unsubscribe from this group and stop receiving emails from it, send an email to gpstest_andro...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/gpstest_android/1e47c0b1-88e2-45e1-aad0-727999be8880n%40googlegroups.com.

[binghy17]

Hi all, the answer from my side is yes.

I've recently collected data with a Xiaomi Mi8 smartphone, processed the raw measurements (cleaning data) with modified Matlab routines coming from GNSSLogger repository, produced a Rinex file, and used it into RTKlib with station measurements I've access to, and I obtained cm positioning. Processing the carrier phase measurements in the most accurate way has not been an easy task, anyway. All these steps have been conducted in post-processing (next step is moving to real-time) and acquiring measurements using a redirection from a survey-grade antenna close to the smartphone (next step is to use only the smartphone antenna).

[Sean Barbeau]

Please let me know if you run across an Android-friendly library to do carrier phase measurements and RTK, preferably licensed under Apache v2 or MIT. Doing this real time on the phone is something I'm interested in it as well.

Sean

--
You received this message because you are subscribed to the Google Groups "GPSTest" group.
To unsubscribe from this group and stop receiving emails from it, send an email to gpstest_andro...@googlegroups.com.

To view this discussion on the web visit https://groups.google.com/d/msgid/gpstest_android/489ead3a-4c02-4ba2-8d70-78e9c1a84047n%40googlegroups.com.

[Louis Philippe]

I am interested too...

Sent from my iPhone

On Jul 20, 2022, at 8:55 PM, Sean Barbeau <sjba...@gmail.com> wrote:



To view this discussion on the web visit https://groups.google.com/d/msgid/gpstest_android/CAEkVq%2BiG4z8iGCJH6NbyomwotW1G-19jPiCcPOpDa2RPRHYpxQ%40mail.gmail.com.

[binghy17]

For the moment, it's all done inside Matlab in post-processing.

Steps:

- acquiring raw measurements with GNSSLogger app (using the signal coming from a survey grade antenna and redirected to the smartphone so the C/N0 is high enough);

- cleaning/processing the raw measurements through Matlab routines coming from GNSSLogger app repo, opportunely modified to process data in the right way;

- generating a 3.03 Rinex file in Matlab with all the measurements;

- using the Rinex file with RTKlib to obtain accurate positioning, similarly to what GEO++ app does.

Problems encountered with GNSSLogger app:

- recording measurements and Rinex from the app at the same time was going to fail (freezing the app). I've noticed that this problem has been solved;

- using Rinex coming from GEO++ app (into RTKlib): ok, similar results;

- using GNSSLogger Rinex file (into RTKlib): no fix, only float. Still deserve to be solved.

Next steps from my side:

- tests using only the antenna inside the smartphone (open sky scenario);

- real-time transition: this will let me use some library for sure and programming an app, I'll check and let you know. This will take time, though;

- adding more algorithms to process measurements smoothly.

[Miguel Faro]

Hello, im looking into a similar but even more budget oriented alternative using an esp32 board and a simpler gnss module ( UC6580) that doesnt provide raw gnss data. I dont require cm level accuracy but rather a middle ground between standalone gnss and full rtk with raw gnss data. What kind of accuracy do you think i would get (i was aiming for sub meter). Also i would like to confirm that is even possible to apply rtk corrections to simple NEMA data and that there is an advantage over standalone GNSS.

Thank you very much, any help is greatly apreciated.