Hi,
This is the subtitles extract  from a movie of a recent flight with a PH4-RTK using local RTK corrections. At the time of recording, we know that RTK correction quality was very good.

300
00:04:59,000 --> 00:05:00,000
F/5, SS 500, ISO 100, EV 0, RTK (45.289390, -12.770514, 33), HOME (45.287012, -12.769161, -19.49m), D 298.72m, H 40.08m, H.S 0.00m/s, V.S 0.00m/s, F.PRY (-0.2°, 1.4°, -178.7°), G.PRY (-90.0°, 0.0°, -178.7°)

301
00:05:00,000 --> 00:05:01,000
F/5, SS 500, ISO 100, EV 0, RTK (45.289390, -12.770514, 33), HOME (45.287012, -12.769161, -19.49m), D 298.72m, H 40.09m, H.S 0.00m/s, V.S 0.00m/s, F.PRY (-0.6°, 1.2°, -178.9°), G.PRY (-90.0°, 0.0°, -178.7°)



From our notes, we know that the drone was around 20.5m RTK as we saw on the remote control.


I don't understant why we have an RTK altitude of 33 in those lines.


It seems that the RTK altitude could be equal to alt_rtk_HOME + H  ie -19.49m+H 40.09m = 20.6m
But Why 33 in this RTK field ?
Anyone knows something ?
Thanks

45N 12W are coordinates somewhere in Atlantic Ocean. Don't you think there is another problem here ?

LV_Forestry Posted at 9-11 07:13
45N 12W are coordinates somewhere in Atlantic Ocean. Don't you think there is another problem here ?

it should be read as 45.289390W, -12.770514S
Those values are extracted from the movie. I did not created them !

IPGP Posted at 9-11 07:25
it should be read as 45.289390W, -12.770514S
Those values are extracted from the movie. I did not created them !

Somewhere in Brazil at 750m altitude ?

LV_Forestry Posted at 9-11 07:29
Somewhere in Brazil at 750m altitude ?

Sorry it should be read as
-12.770514, 45.289390
https://maps.app.goo.gl/6KJVGUZUauUxqwRa6

How do you find 750m ?

IPGP Posted at 9-11 07:37
Sorry it should be read as
-12.770514, 45.289390
https://maps.app.goo.gl/6KJVGUZUauUxqwRa6

750m is the SRTM value at -12.770514, 45.289390 (12S 45W)

I'm not asking you this to annoy you or lecture you. But before getting interested in altitude accuracy, there are a few things to look at closely.

First of all, what are we talking about? WGS coordinates, is that correct?

The standard is latitude (which hemisphere),

Positive = North
Negative = South

then Longitude.
Positive = East
Negative = West

As you can see there is already a first problem of normalization of the writing of the coordinates. It is DJI material, so far nothing abnormal.

Now that we know where we are, we can already identify whether the Home point is good.


According to SRTM the point must be at an altitude of 13m AMSL.
Then we will convert the altitude of this point to that of the WGS84 system. To do this we take the value of the SRTM and add that of the EGM96 which is the model (geoid) that allows the conversion. I go into detail here:
DJI Products EXIF Wrong Altitude / How to solve it | DJI FORUM




The value is -16.01m, which means that the WGS altitude of the starting point is 16m below the SRTM altitude (AMSL assimilated)
13m -16m = -3m

Your home point should logically indicate -3m.
Unfortunately in the data of the SRT file we can read -19.49m. There is therefore an offset of approximately -16m

33-16 = 17m (true WGS altitude of the drone)

from -3m to 17m = 20m! So it stick to the declared 20.5m Admit that nature is still well made!

Now that we know everything is fine, I have just two question left to ask you.
1/What is your local base station?
2/How did you install it?

I expect answers like:
1/DRTK-2
2/Well I put it in a random place and started it up and that's it

LV_Forestry Posted at 9-11 08:11
750m is the SRTM value at -12.770514, 45.289390 (12S 45W)

I'm not asking you this to annoy you or lecture you. But before getting interested in altitude accuracy, there are a few things to look at closely.

Thanks for all your explanation !

Our RTK provider is https://www.reseau-teria.com/ and the local base station we used is in the same small island where we flew. The green dot on this map

which is  around 3.7km away.

What I am interested in is the absolute altitude and not really the distance between the drone and the ground.
Thanks

IPGP Posted at 9-12 07:15
Thanks for all your explanation !

Our RTK provider is https://www.reseau-teria.com/ and the local base station we used is in the same small island where we flew. The green dot on this map

Then we have a bigger problem than expected.
Let me time to think about it and doing some test.

LV_Forestry Posted at 9-12 07:19
Then we have a bigger problem than expected.
Let me time to think about it and doing some test.

@LV_Forestry Did you have time to do some tests ?
Thanks

IPGP Posted at 10-3 22:42
@LV_Forestry Did you have time to do some tests ?
Thanks

Not yet, to be honnest i totaly forgot about that.
This week end i will do photogrametry, i will try to remember to take a video.

LV_Forestry Posted at 10-3 23:35
Not yet, to be honnest i totaly forgot about that.
This week end i will do photogrametry, i will try to remember to take a video.

So I made an interesting test outdoor today

• with a DRTK-2; a PH4RTK remote controler with an RTK provider in FIX mode, I recorded the position of a point on a terrace. The absolute altitude of this point is 101,8144403 (I remove the height of the DRTK-2 antenna mast)
  

With my PH4RTK that I put on top of the previous point (not flying..)  I recorded @ pictures and a movie :

• for both picture 1 and 2 :
• RTK flag is 50 (excellent RTK correction)
• Absolute Altitude               : +101.84
• Relative Altitude               : -0.00
• for the movie I have
• Home point altitude = 102.04m
• While recording the movie the PH4RTK remote indicate a RTKH of 102.0m
  

I can estimate that the drone camera is 5cm above the point I measure with the DRTK-2. 101,8144403+0,05 => 101,8644403

So what can we conclude ?

DRTK-2 measurement and Absolute Altitude of pictures are the same
DRTK-2 measurement and home altitude of the movie have around 20cm difference
DRTK-2 measurement and RTKH in the PH4RTK remote have around 16cm difference

Any idea ?

IPGP Posted at 12-12 07:24
So I made an interesting test outdoor today

with a DRTK-2; a PH4RTK remote controler with an RTK provider in FIX mode, I recorded the position of a point on a terrace. The absolute altitude of this point is 101,8144403 (I remove the height of the DRTK-2 antenna mast)

We can conclude that the altitude in video is shortened as to not have so many numbers after the coma. Or, its not RTK altitude but kind of initial altitude+ relative.

I stil haven't managed to get a P4R to try. I tried with P4M, it doesn't record SRT, so nothing helpful there.

LV_Forestry Posted at 12-12 08:35
We can conclude that the altitude in video is shortened as to not have so many numbers after the coma. Or, its not RTK altitude but kind of initial altitude+ relative.

I stil haven't managed to get a P4R to try. I tried with P4M, it doesn't record SRT, so nothing helpful there.

Here is the data in the movie. I don't know what 25 represents after the drone poistion.

IPGP Posted at 12-13 03:46
Here is the data in the movie. I don't know what 25 represents after the drone poistion.
[view_image]

I don't know. And of course in the manual there are no reference.

Have you tried to write to enterprise@dji.com ?

IPGP Posted at 12-13 03:46
Here is the data in the movie. I don't know what 25 represents after the drone poistion.
[view_image]

Maybe an expression of accuracy in mm