Please select Into the mobile phone version | Continue to access the computer ver.
50m Base Coordinate Tolerance Is a Bad Thing™
1388 16 2019-4-5
Uploading and Loding Picture ...(0/1)
o(^-^)o
patiam
First Officer
  • >>>
United States
Offline

Today we were unable to fly using the D-RTK 2 at our regular flight testing zone because the base station coordinate setup would not accept our known coordinates. It repeatedly rejected them as being > 50m from the base's own solution.

This is for a benchmark in which we have extreme confidence, that has been shot independently and repeatedly using a Trimble 5700 and NetR5 RTK with rapid-static processing by OPUS, with a Total Station, and has been used for calibration and as a GCP or Checkpoint in countless eBee RTK flights, dozens of non-RTK P4 flights, and at least half a dozen P4P RTK flights. We're damn sure we know the XYZ of this spot to within a couple cm (or better). And the D-RTK 2 has accepted those very coordinates on many occasions previous to today. We've had it briefly reject them in the past, but have always been able to get it to take them after another try or two.

We tried multiple times to enter the cordinates, and cycled the power on the base as well as the A/C. When we still couldn't get past the > 50m rejection fail, we switched to using NTRIP from our own nearby Trimble NetR9 base station so we could get the needed flights in and to let the D-RTK 2 run for a while on its own in hopes it would realize where it really was and eventually accept our coordinates.

After two short NTRIP flights (maybe 20 minutes total) we tried the D-RTK 2 again. Same issue. Nothing we could do would make the app accept the true coordinates. I believe we recorded the base's own solution either by writing it down or taking a pic of the screen, but don't have those with me to compare yet.

We ended up doing another NTRIP flight to complete what testing we could. But part of what we wanted to test was performance using the D-RTK 2.

I do know the D-RTK 2 and A/C were last used together about 15-20 km from our test zone, so maybe it was remembering those coordinates, or maybe something else was wrong. But wahtever the case, the coordinate rejection logic is a bad thing, for a number of reasons.
2019-4-5
Use props
HTA_RP-1
lvl.4

United States
Offline

Confidence in the equipment and workflow is critical.  When equipment does not work consistently when following the same workflow that has previously worked, doubt begins and doubt leads to looking for alternate consistent reliable solutions....
2019-4-5
Use props
patiam
First Officer
  • >>>
United States
Offline

HTA_RP-1 Posted at 4-5 13:46
Confidence in the equipment and workflow is critical.  When equipment does not work consistently when following the same workflow that has previously worked, doubt begins and doubt leads to looking for alternate consistent reliable solutions....

yeah luckily in this case we had a fallback and the mission wasn't critical, just some testing and teaching right at home.

but if we were far afield where there was no cell coverage on an important and costly mission, it would have been much more of a problem.

as you noted, my confidence in the robustness of the system is waning.
2019-4-5
Use props
patiam
First Officer
  • >>>
United States
Offline

So on Sunday a colleague of mine tried using the D-RTK 2 again at a different location, and it would not accept the known coordinates for that benchmark either. And interestingly, the base's own self-calculated position only differed by about 2.3m from the coordinates he was entering [off by a couple (10^-5)°. ]



So something is broken. Either there is a some internal memory holding a previous set of coordinates (that isn't displayed on the screen), GS RTK isn't doing the math correctly.

Turns out that he wouldn't have been able to fly even if the D-RTK 2 was working, as he was also shut down by geofencing anyway, which is another very Bad Thing™.

2019-4-8
Use props
jivago
Second Officer
Flight distance : 23671 ft
Iran
Offline

patiam Posted at 4-8 09:45
So on Sunday a colleague of mine tried using the D-RTK 2 again at a different location, and it would not accept the known coordinates for that benchmark either. And interestingly, the base's own self-calculated position only differed by about 2.3m from the coordinates he was entering [off by a couple (10^-5)°. ]

[view_image]

If I were you, try to find out the fixed coordinates of D-RTK and after finishing my flight, execute the shift between my coordinates and D-RTK's coordinates.
Because unfortunately there's no evidence that DJI company follows this forum and wants to solve these problems.
2019-4-8
Use props
patiam
First Officer
  • >>>
United States
Offline

Thanks @jivago...

I can come up with a number or workarounds, but currently I'm trying to make the system work as advertised. It seemed like it was doing that for a handful of flights, but now, not so much.

I'm fully aware that actual support fro DJI is not to be had on this forum. I post my experiences, both successful and not, in the hopes it might help other users and that others might help me.

2019-4-8
Use props
patiam
First Officer
  • >>>
United States
Offline

Tried again at a different BM with same results. Precise known coordinates rejected as being ">50m from current position" despite the fact that they were more accurate and only a couple m from the D-RTK  base's calculated solution.

If this doesn't get fixed the system is garbage unless NTRIP/VRS is used. There's no straightforward way to collect data on an unknown point and post-process, and it won't accept known coordinates. Yes, there are PPK and other workarounds, but that's not what DJI is marketing the P4P RTK as. And vendor support is dismal. They're just as (or more) uniformed as users. We've figured out literally every question we've posed to a vendor before they even get back to us with a response of "we're not sure, we'll ask DJI".

In case it's not obvious, we're pretty frustrated at this point...
2019-4-9
Use props
patiam
First Officer
  • >>>
United States
Offline

Looks like a firmware mismatch between base and AC may have had something to do with the ">50m" issue. Seems to be accepting coordinates now after updating. Will test tomorrow afternoon with several consecutive short flights at our calibration site.
2019-4-22
Use props
HTA_RP-1
lvl.4

United States
Offline

patiam Posted at 4-22 10:08
Looks like a firmware mismatch between base and AC may have had something to do with the ">50m" issue. Seems to be accepting coordinates now after updating. Will test tomorrow afternoon with several consecutive short flights at our calibration site.

That is good to hear!
2019-4-23
Use props
patiam
First Officer
  • >>>
United States
Offline

Thanks, although its a bad system if out of date or mismatched firmware can cause that sort of problem. It was working, and then it wasn't. Now, hopefully, it is working again.

I'll let you know how it goes after our afternoon test flights today. Best case scenario our bad Z accuracy on our last flights was due to the firmware problem as well.
2019-4-23
Use props
LYVFD8142
Second Officer

United States
Offline

I'm having the same issue and I have everything updated. I think it's a bug on DJI's side.

Here is the test, I was able to set the Lat/Long/Elevation (awhile back at a site and it took the info). The past few days I'm flying 50 miles away and it won't let me put in the new Lat/Long/Elevation, not even the Lat/Long/Elevation of what the Base Station is reporting (now that's funny). You can't remove the old Lat/Long/Elevation either which you should be able to do which I would expect when they are blank they would use the reported Lat/Long/Elevation from the RTK Base Station itself.

I will go back to the original location where I was able to set the Lat/Long/Elevation and see if I can remove that from the unit.
2019-6-15
Use props
patiam
First Officer
  • >>>
United States
Offline

That this is still happening to folks is really worrisome. I had hoped this was all fixed with recent firmware updates.

This system is worthless as a professional mapping tool until we can have confidence that when we arrive at a new site, no matter how far it is from the last place we used it, we can enter known coordinates for the D-RTK 2 base station and have them accepted so we can fly.

There is no need for a lockout based on difference between entered coordinates and either the current or any previous position solution. To have one is ridiculous. At worst this should trigger a dismiss-able popup. Or nothing.
2019-6-15
Use props
LYVFD8142
Second Officer

United States
Offline

I agree, I think there should be a Clear Data Button that removes all 3 data (Lat/Long/Elevation). And then no matter what, Signal or not, a User Input should always override the what the RTK unit thinks data if set.
2019-6-16
Use props
mbuonanno
lvl.4

Italy
Offline

patiam Posted at 2019-6-15 20:47
That this is still happening to folks is really worrisome. I had hoped this was all fixed with recent firmware updates.

This system is worthless as a professional mapping tool until we can have confidence that when we arrive at a new site, no matter how far it is from the last place we used it, we can enter known coordinates for the D-RTK 2 base station and have them accepted so we can fly.

Hi patiam,

did you solved the ">50m" issue and how?  Have you encountered the problem after the last firmware update (10/03/2020) ?

The D-RTK2 user guide (v.1.2) at page 9 says:
"If the D-RTK 2 Mobile Station is restarted after input coordinates are successfully set, these coordinates will be used only if the difference between the actual coordinates and the set coordinates is less than 5 m. Otherwise, the actual positioning coordinates will be used."

Do you think the ">50m" issue could be related with the above sentence (e.g. it is not possible to reset the coordinates previuosly sent by the GS RTK app to the DRTK2 base) ?

Cheers,
Maurizio
2020-3-29
Use props
patiam
First Officer
  • >>>
United States
Offline

mbuonanno Posted at 3-29 10:39
Hi patiam,

did you solved the ">50m" issue and how?  Have you encountered the problem after the last firmware update (10/03/2020) ?

Hi Marizio-

I haven't been involved with flying our P4 RTK since last posting in this thread, so unfortunately I can't comment on how it is working now or even if we've updated to the most recent firmware (I think not yet).

I do think that both of DJI's coordinate reality check failsafes (the 5m and 50m ) have no place in a professional survey grade kit other than perhaps to trigger a dismissable pop-up warning. The coordinates entered by the user should be honored no matter what, that's the whole reason the option exists.

Hope all is well in Italy. Aspetta qui.
2020-3-29
Use props
mbuonanno
lvl.4

Italy
Offline

patiam Posted at 3-29 10:53
Hi Marizio-

I haven't been involved with flying our P4 RTK since last posting in this thread, so unfortunately I can't comment on how it is working now or even if we've updated to the most recent firmware (I think not yet).

Hi patiam,

thank you for your quick replay.

In Italy things seams to go better than the last week. Our doctors, hospital staff menbers and voluntiers are doing an incredible job. Most of us are obbliged to work at home to avoid spreading  the coronavirus infection.

I am working at home and I am using my time to organize the whole set of informations about the DJI Phantom RTK that are available in the DJI Forum.  If i will do a good job, I will share on the forum the document for revision and amendments. :-)

I wish you and people in your country to exit as soon and safe as possible from this difficult experience.

Maurizio
2020-3-29
Use props
patiam
First Officer
  • >>>
United States
Offline

mbuonanno Posted at 3-29 13:57
Hi patiam,

thank you for your quick replay.

Hi Maurizio-

Please do share the product of your efforts to organize and summarize P4 RTK info- it will be helpful to many I'm sure. The forum thread format is good for figuring out things in real time but not as good as a reference resource...

Thank you for your kind wishes regarding the virus. We are just beginning to see how it will really affect us here, beyond having to stay home and of course the toilet paper shortages *smh*. I'm afraid it will get much worse before it gets better but we will persevere.

Ganbarou!

-p
2020-3-30
Use props
Advanced
You need to log in before you can reply Login | Register now

Credit Rules