Maintain FIX during movement with M+

Hi @Tomlogan,

Do I understand correctly that you’ve configured to transmit the following RTCM3 messages?

  • 1002 GPS L1 observations
  • 1006 ARP station coordinate

@Tomlogan,

It’d be great if you could just generate the Full System Reports on both units and send them to me in PM. The information in the reports will help us see the settings as well as determine the cause of the issue.

HI @polina.buriak,
Yes I configure for GPS and ARP, I may have forgot GLONASS, is it the cause of the instability ?
I will send you the system report

Hi @Tomlogan,

The information about enabled RTCM3 messages will help me check if there’s anything missing in the data received on the rover. After it’ll be easier to understand what could go amiss.

Looking forward to receiving the Full System Reports.

Hi @Tomlogan,

Thanks for sharing the Full System reports!

I’ll delete your message as the report may contain sensitive information. We’ll look into it and I’ll get back to you with news.

Regarding your questions, the rover position is relative to the base position. It means that if you establish the base’s coordinates with the Average Single, it’s absolute accuracy is meter-level. So is the absolute rover’s.

If you require absolute centimeter-accuracy on the rover, you need to know the base coordinates with the centimeter-level accuracy as well. For that, as you have mentioned, you can use the NTRIP corrections.

Hi @polina.buriak,
Hope you are doing well with this exceptional circumstance.
Any news about the system reports ?

Hi @Tomlogan,

Thank you, hope you’re well too!

Currently, we’re looking into the issue with the fix solution. I’ll inform you once there’s any update on it.

Hi @Tomlogan,

We’ve just released the v2.22.4 firmware version. This update contains an important fix that should improve the solution in RTK.

I’d ask you to update your receivers to this latest stable release and check how it works for you. Please keep us posted on the results.

Hi @polina.buriak,
I got good results with the new firmware, got a FIX with my rudimentary setup.
It seems more stable than before. I can’t test it on the final setup (on a USV) but I think it will be OK.

IMG_20200424_162536_s IMG_20200424_162527_s

Open sky with no obstacles in a field

Here are the logs
raw_202004241414_UBX.zip (2.5 MB) solution_202004241414_LLH.zip (105.5 KB) base_202004241414_RTCM3.zip (205.4 KB) SystemReport_rover.zip (238.6 KB) raw_202004241405_UBX_base.zip (876.3 KB) SystemReport_base.zip (275.0 KB)

Does it need improvements on the rover and the base ?

Hi @Tomlogan,

Thanks for the test! It seems like the latest update is working well for you.

However, I can see that you use GPS and GLONASS only. It means that both base and rover simultaneously see only 5 satellites with SNR around 45 (G18, G21, G26, G31, R09). I’ll attach the screenshots of the logs below.

This is the rover log:

This is the base log:

This might cause some difficulties for the rover to obtain the fix. So I’d recommend enabling more satellite systems for your tests so that the rover can use more data for the calculations. This will help increase the stability of the Fix solution. Our default recommendation is all satellite systems (GPS + GLONASS + Galileo + SBAS + QZSS) enabled at 1 Hz for the base and at 5 Hz for the rover.

It’d be great to see your results with the final setup and on the moving rover!

Hi @polina.buriak
We have order a Reach RS+ as base, hope it will more practical to “set and forget”
I will try again with your configuration.

1 Like

Elevating the base will also help quite a bit. I know it is hard on ground based drone, but it can have a huge impact.

Sure that’s why we order a RS+ with a tall tripod :slight_smile:

Sorry, meant both units… so also your rover, most of the time, especially the rover, as it is usually the one facing difficult conditions.

Hi @polina.buriak and @wizprod
I test the reach rs+ with the previous rover M+. After a period of stabilization, it works fine and i got a fix on the rover.

My setup

Here are the logs
SystemReport(2).zip (287.8 KB) solution_202005181252_LLH.zip (139.8 KB) raw_202005181252_UBX.zip (3.8 MB) base_202005181252_RTCM3.zip (437.9 KB) SystemReport(1).zip (244.2 KB) solution_202005181246_LLH.zip (52.2 KB) raw_202005181246_UBX.zip (1.7 MB)

Do you see anything to improve ? If not I can close the thread.

Hi @Tomlogan,

Sorry for the delayed response!

I can see that it takes a rather long time for the rover to establish a Fix solution. I’ll attach the screenshot below.

It seems like it happens because of the low observational data quality on the rover. You can see that there are not many satellites with SNR over 40. Also, there is a significant amount of the cycle slips present.

This may be due to the setup specifics. From the photo you uploaded, I suppose that Reach M+ is placed on the ground near the case for the test. You might want to lift it above the ground to improve the satellite reception and exclude the possibility of LoRa signal interruptions from the obstacles. I believe that with the improved rover placement it’ll be easier to achieve fix.

It’d be cool to check your further results :slight_smile:

Thanks for your answer, the M+ will be on top of a boat
It will be better I think. I will tell you once I go back on it.

BTW how can we display the logs as graph like yours ?

1 Like

@Tomlogan,

You can open both observational files and position logs with RTKPlot from the RTKLib QT apps. You can download them in our GPS Post-processing guide. More information on how to check the data quality can be found in Understanding PPK solution and analyzing logs from Reach guide.

Hey there,

Just wanted to add that our latest stable version v2.22.5 should shorten the Fix reinitialisation time and improve the RTK solution even more. It’d be great if you could confirm that it works well for you.

You can check out the full list of fixes and improvements in this community forum thread.

1 Like

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