120ccpm Posted at 2-2 09:32
To the OP: have you gone back to what you learned is the real crash site, and tried to retrieve the AC? Or is it in a completely unaccessible location?

@120ccpm

... as I had repeatedly emphasized, after analyzing the txt file (flight data) using "AirData" it was clear to me where the MM had crashed ...

Here in this forum I wanted to find out how it could be that the "Find My Drone" function could display such a completely absurd position (PDrone).
slup has solved this riddle for me, I am very happy about it!

Regarding your question:
The bad weather, some with some snowfall, the difficult terrain at the crash site and - especially of course - the lack of any support from the MM the next day and the following, all of which would have made it a most likely unsuccessful search, and maybe a very dangerous one ...

Immediately after the crash, everything was different, the chances would have been many times better ...

Frank

frankymusik Posted at 2-2 13:57
@120ccpm

... as I had repeatedly emphasized, after analyzing the txt file (flight data) using "AirData" it was clear to me where the MM had crashed ...

I totally understand that your concern was the incorrect position reported by Find My Drone after the crash.
I was just asking if - after learning where your MM actually crashed - you went back and tried to find it.
It could be a fun adventure, and definitely a nice story to report on this forum.

120ccpm Posted at 2-2 14:45
I totally understand that your concern was the incorrect position reported by Find My Drone after the crash.
I was just asking if - after learning where your MM actually crashed - you went back and tried to find it.
It could be a fun adventure, and definitely a nice story to report on this forum.

@120cpm

I know, I know, I know ...

But, at the moment it is really too dangerous if, despite everything, you do not know exactly where to look ... The area is simply too large for a targeted search, especially since you have to suspect that after the crash (and GPS failure --- if you study the flight data) the possible 'landing site' is still rather uncertain ...

It would be interesting for me (to complete the incident):

• DJI finds the error and avoids displaying such nonsense in the interest of all users
• DJI recognizes that this behavior of the MM (Find My Drone) meant that an immediate search had to be unsuccessful, and "Care Refresh" should therefore apply ...
  

Below is another picture of the landing site, in my view ... (of course not so green at the moment)

Frank

slup Posted at 1-31 05:17
The explanation for the position shown in "Find my drone" lays in the difference between the 2 log types, TXT & DAT.

The TXT log is generated from motor on to off ... but the DAT log is from power on to off (off state also occurs if the connection with the RC is broken).

@slup

May I please ask an additional question after you have largely clarified the causes of the incorrect display of the landing site ("Find My Drone" function), please?

The last known GPS position in the DAT file is quite clear:
          50.9525591° 14.0725344°
is that correct?

However, the position displayed in the "Find My Drone" was:
         Pdrone = 50.948902° 14.072694°
I am very sure about that!

There is another 430m (1,400 ft) between these two positions (see the picture below, the red route was also "invented")!

Who or what "extrapolates" these values ​​in the flight data so that Pdrone is finally displayed?

So there has to be an algorithm behind it that tries to predict the landing site "better" ...?
That obviously goes completely wrong, doesn't it?

DJI certainly has a lot to do here, doesn't it?

When the software "learns" that the motors are stopped etc. pp. (a lot of information is still available!), then no position should be issued that is approx. 1,400 m away from what was last known ... (crash site)

Is that perhaps an argument for making DJI a goodwill ... ???

Frank

frankymusik Posted at 2-2 13:57
@120ccpm

... as I had repeatedly emphasized, after analyzing the txt file (flight data) using "AirData" it was clear to me where the MM had crashed ...

Hi I have simulated your loss in google earth.

Flight View

Camera View

It hit a tree on RTH and then tumbled down a cliff.

grahamjohnson10 Posted at 2-3 11:02
Hi I have simulated your loss in google earth.

Flight View https://youtu.be/ojK2fuVU-i4

Fantastic simulation!  Great job, really!

Frank

frankymusik Posted at 2-3 01:08
@slup

May I please ask an additional question after you have largely clarified the causes of the incorrect display of the landing site ("Find My Drone" function), please?

I would suspect that the difference with the real location and the location advised by find my drone was due to the tumbling of the drone after impact.

The tumbling would have caused the GPS lock to fail. Giving a high degree of error.

For additional information be warned that flying below takeoff hight near a cliff is going to cause a significant problem.

The accuracy of the GPS lock is determined by the number of satellites in the lock and their location in the sky. The accuracy of the lock is its HDOP. (Horizontal Degree of Precision.) Not by the number of satellites in the lock. When you took off, you had a full sky available and would have had a low HDOP. The lower the better. At a cliff edge the view of the sky is significantly reduced. The HDOP will increase and could be as large as a 1klm radius. At best lock its about 2 metres.

So basically do not fly close to cliff edges. If you do set the RTH to maximum height to ensure that the RTH will rise to an altitude where more of the sky is visible and the HDOP lower.

Also for diagnostic purposes ignore the GPS altitude. The VDOP (Vertical Degree of Precision.) Cannot be trusted it's a very course measure, the reason that all drones have a barometric altimeter.

If the drone reports the HDOP back to the log / application then DJI could include a radius to assist you to find where it went down. e.g. 10 metres radius, or 1klm radius. That would have maybe, saved you from your trek to find a bridge over the river.
  
Cheers Brian

bjr981s Posted at 2-3 22:31
I would suspect that the difference with the real location and the location advised by find my drone was due to the tumbling of the drone after impact.

The tumbling would have caused the GPS lock to fail. Giving a high degree of error.

@bjr981s

A very interesting addition to the information available so far, really!

But if, as slup explained in Post #63, the GPS positions are saved in the TXT file (of the flight data) while the motors are still powered (GPS positions which are all very plausible!) ...
... then why generate further GPS positions (in the DAT file) that make no sense?

And why at the end (in the "Find My Drone" function) generate an additional GPS position that is not even contained in this DAT file?

Can you really justify that? Does it make any sense in any situation? (see my post #165 too)

For me, these are still elementary mistakes that DJI should just avoid ...

Frank


PS:
Your comment ('... be warned that flying below takeoff hight near a cliff is going to cause a significant problem') is understandable, but nowhere during this flight did I drop below this take-off altitude. The MM flew during the entire flight in the RTH height of 30m above "zero" ... (= height of the starting point)

frankymusik Posted at 2-4 00:47
@bjr981s

A very interesting addition to the information available so far, really!

Hi Franky,

Your HDOP and PDOP values from 10s before till 10s after the tree crash.
So up to the impact all GPS and DOP values normal. (logic as fly position in air to sats wasn`t blocked by nothing)

BTW  DOP from normal values in log to "99.99" values, i changed those values to zero to make chart better readible.
cheers
JJB

JJB* Posted at 2-4 02:47
Hi Franky,

Your HDOP and PDOP values from 10s before till 10s after the tree crash.

Thanks JJB, how did you get that graph.

grahamjohnson10 Posted at 2-4 03:45
Thanks JJB, how did you get that graph.

copy paste some values form DAT file columns into Excel,  just  a basic chart.

JJB* Posted at 2-4 04:20
copy paste some values form DAT file columns into Excel,  just  a basic chart.

Thanks JJB

JJB* Posted at 2-4 02:47
Hi Franky,

Your HDOP and PDOP values from 10s before till 10s after the tree crash.

Hi JJB,

there is certainly a lot of information to be extracted from the flight data after the fact.
As a "normal user" one should of course have little to do with it ...

Important rules (such as the correct RTH height) are of course essential!

In spite of all pilot training (at the MM most certainly beginners), the manufacturer in particular has to deliver ... DJI should study all of these analyzes, which were carried out here in the forum.

In my case, it is not the crash itself that worries me. Here I knew from the start that flight errors (specifically the RTH altitude I chose) were responsible.

It is simply the alleged last position that still makes me very angry because DJI obviously cannot evaluate the measured / received data satisfactorily, so sends the user "into the desert" ...

Frank

frankymusik Posted at 2-4 08:22
Hi JJB,

there is certainly a lot of information to be extracted from the flight data after the fact.

Hi Franky,

I understand your point, but in the situation a drone is out of its normal flight (like yours falling down that hill) you cannot expect from the software to give correct values anymore.

So with a normal 'failure', i.e. critical low batt landing far from HP, the find my drone option does a good job. (as long as drone is in LOS, count as well for radio signals)
Proof to be found on this site with other users/flightlog/experience.

Reason i made this last chart to convince your that DJI is not to blame for errors in the find my drone data in your case. Just my 2-cents.

cheers
JJB

JJB* Posted at 2-4 09:18
Hi Franky,

I understand your point, but in the situation a drone is out of its normal flight (like yours falling down that hill) you cannot expect from the software to give correct values anymore.

Finally I have a new MM and Care Refresh is also (again) activated ...

No, DJI was not merciful, and that was never to be expected.

Now I am looking forward to nice, suitable weather to be able to start again.

I've learned what the main problems can be for an accident to happen:

1)  Too low outside temperatures (probably too high too)
2)  High humidity or even sudden precipitation
3)  Wind of any conceivable strength (swirls could occur)
4)  Flight at high altitude
5)  Flight far away
6)  Radio waves of any origin, especially where the MM is currently flying
7)  Batteries that suddenly no longer have sufficient voltage
8)  Propellers that show invisible damage
9)  Flying too fast by using the S mode
10)  ...
.
.
.
n-1)  Incomplete control software of the MM, RC, batteries and / or Fly app
n)  Stupid beginners who cannot set the required RTH level

Probably I have left out many influencing factors, but diligently studying the forum posts will increase the wealth of experience ...

Nevertheless, I am looking forward to the next flights and hope that I will be a lot lucky then, although I hardly deserve it ...

MAVIC MINI - THE EVERYDAY FLYCAM - Everyone can fly

Frank


PS:
Thanks again to all those who have sacrificed their time so that I have a little more courage again ! Thank you!

frankymusik Posted at 2-4 13:52
Finally I have a new MM and Care Refresh is also (again) activated ...

No, DJI was not merciful, and that was never to be expected.

All above is about what not to do, surely knowing what to do is much more useful and positive it seems you lost your drone because you didn’t know what to do in the situation you were in mmmmm.

very interesting post

Maybe somebody found it.
https://mavicpilots.com/threads/ ... i-in-germany.82299/

netzwerg Posted at 2-18 05:50
Maybe somebody found it.
https://mavicpilots.com/threads/found-mavic-mini-in-germany.82299/

Another lesson learned.  Put an "If found please call" sticker on your drone.

BobWinNV Posted at 2-18 12:04
Another lesson learned.  Put an "If found please call" sticker on your drone.

or a Readme.txt on the SD card, with an email adress

He ate a humble pie... fingers crossed they found his drone.

https://blog.saechsische-schweiz.de/regional/sonstiges/luftschauen-drohnen-quadrocopter-octocopter/

I cannot understand, he's got a new MM w/ Care Refresh without bringing in his old drone ? Btw did he acknowledge the find ?

Question:

In the dji go4, there is a switch for correcting the gps map position in mainland china.

That is in the fly app, too, I assume.
And perhaps switched on and was "correcting" the correct GPS position?  
So that the find my drone is showing a wrong position, way of course?

Just a thought

hallmark007 Posted at 2-1 11:41
Once your drone moves it will reference its height to that of ATO at take off, so if it flys over a ravine below homepoint it should register a minus figure , so 30m over homepoint fly forward 10m over the ravine with ground level 200m below you craft should register - 230m and so will change this as it moves on its journey, labroids tried to emphasize this on so many occasions but was continually shot down by all who were insisting on using ASL and AGL which were not needed to review this log .

I'm really getting into this thread late...Aug 4...but what you say here can be confusing to a newbie...unless you're talking about the logs and not the altitude displayed in the app. If you're talking about altitude displayed in the app and you are 30 meters over the HP (ATO as you state) and fly 10 meters forward out over the ravine whose bottom is 200 meters lower, your altitude on the app will still show 30 meters...not -230 meters as you state.  Now, if you descend into the ravine and go below the HP altitude (ATO), you will start to see the altitude in the app go to negative numbers (i.e. the meters below the HP altitude).  On the otherhand, if you fly straight forward and never change the throttle (left stick forward or aft) the app altitude will not change just because the terrain under the drone changes.  There is no Radio Altimeter and the drone does not use GPS to determine terrain height below. Again, I might have misunderstood what you were referencing.  I just didn't want any newbies confused.

frankymusik Posted at 2-4 13:52
Finally I have a new MM and Care Refresh is also (again) activated ...

No, DJI was not merciful, and that was never to be expected.

Did you ever see Reply 179???   Was that your MM???

Rustic17 Posted at 8-5 12:31
Did you ever see Reply 179???   Was that your MM???

No!
I've never seen this before ...      

frankymusik Posted at 8-5 21:53
No!
I've never seen this before ...

Wasn't it in the same area you were flying???  I looked up Rathen on Google Earth and it looked like the same area.

Rustic17 Posted at 8-5 21:55
Wasn't it in the same area you were flying???  I looked up Rathen on Google Earth and it looked like the same area.

Rathen, of course!
Thank you so much!
I registered with "MavicPilots" today and am trying (so late!) to get in touch.
It would be almost a miracle ...

frankymusik Posted at 8-5 22:44
Rathen, of course!
Thank you so much!
I registered with "MavicPilots" today and am trying (so late!) to get in touch.

Hope it works out for you!!!  Keep us posted...

Rustic17 Posted at 8-5 22:46
Hope it works out for you!!!  Keep us posted...

... in any case!

Maybe I will get an answer from there, but that late, 6 months later.