Hello,

Today I lost my brand new DJI mini 3 pro.
I believe I have flown it too high and with some wind, the drone and controller lost connection and the drone didn't manage to return home.
I have the last known coordinates, but I guess this were the coordinates before the signal was lost and I am not sure if it really crashed there because according to the flight log, the altitude was around 60 meters.

Can someone help me read the flight log and try to guess where it could have crashed?
Here is the link to the flight log on phantomhelp and I have .txt, .klm and .csv files if needed.

https://www.phantomhelp.com/LogViewer/8QZ5P3A3OLW7OTVU9IJM/

Thank you very much, I really hope to find it.

This is possibly going to be an interesting thread, you might like to post it over on MavicPilots too.
Question for the likes off JJB and Labroides, after 18:12 can the indicated speeds be relied upon?

I believe I have flown it too high and with some wind, the drone and controller lost connection and the drone didn't manage to return home.

There is a big problem with solving this flight mystery.
At 18:10, there was an unusual interruption to GPS.
The drone was still reporting a high number of satellites, but the flight controller had no confidence in the GPS location data and stopped calculating the drone's location.
For the last 3 minutes of the flight, we do not know where the drone was or what speed it would have been moving in response to joystick inputs.

I'll look into it further and update with what I can work out.

Thank you for your analysis.
There is some kind of transmission line in the vicinity. According to the data, is it possible the drone crashed into it and lost power?

Mamelo Posted at 10-4 15:00
Thank you for your analysis.
There is some kind of transmission line in the vicinity. According to the data, is it possible the drone crashed into it and lost power?

See my updated post above

Maybe it has something to do with me turning off the controller and turning it back on before searching for the drone location (I know now it was a mistake but hopefully I will learn for next time).

Mamelo Posted at 10-4 15:09
Maybe it has something to do with me turning off the controller and turning it back on before searching for the drone location (I know now it was a mistake but hopefully I will learn for next time).

Did you still have signal when you turned off the controller?
Or was signal lost before you turned off the controller?

Labroides Posted at 10-4 15:18
Did you still have signal when you turned off the controller?
Or was signal lost before you turned off the controller?

The signal was lost while the controller was still on

It was strange, the drone was hovering and all of a sudden it started flying and there was a wind warning on the controller before it lost signal

I really enjoy watching Labroides work these things out!  It is like an episode of CSI!

TonyPHX Posted at 10-4 15:27
I really enjoy watching Labroides work these things out!  It is like an episode of CSI!

I don't enjoy the tough ones like this.
It's a real puzzle with too many pieces missing.

I wish I had more info, but I am unfortunately still very new to flying drones. If I can somehow extract more info from the controller, please let me know

Mamelo Posted at 10-4 15:25
It was strange, the drone was hovering and all of a sudden it started flying and there was a wind warning on the controller before it lost signal

Here's where I'm at so far....

GPS was lost at 18:10 with the drone 150 metres west of the homepoint and 87 metres up.
30-40 seconds before this, the drone was higher (113 m) and easily holding position with minimal pitch and roll angles which is a good indication that winds were very light at that height.
It doesn't look like the drone was blown away by strong wind, but may have drifted slowly without GPS.

The homepoint was the highest land in the area, so there were no obstacles to block signal or crash into.

GPS position and speed data are false for teh last 3 minutes of the flight.
Altitude data as well as pitch, roll,compass and joystick data are still good.

Normally a drone in atti mode will land when signal is lost as it cannot RTH.
But the drone in this incident was still reporting itself as in Normal Mode, but not using GPS data, so it might (?) have remained hovering.

The drone was relatively close to the homepoint with good signal strength, so the loss of signal suggests loss of power, although it seems an unlikely coincidence to lose power minutes after losing GPS.

Pitch and roll data at the end of the flight don't indicate a collision with a powerline or other obstacle.
Given the evidence of light winds when data was reliable, I suspect the high wind warning is false.

Labroides Posted at 10-4 15:47
Here's where I'm at so far....

GPS was lost at 18:10 with the drone 150 metres west of the homepoint and 87 metres up.

So no strong winds, no collision with a powerline and no changing the home location.
What other possible causes are there?
Battery life?

Some kind of deliberate interference with the signal?
I was flying near a cultural site although not blocked on maps.

Mamelo Posted at 10-4 15:59
So no strong winds, no collision with a powerline and no changing the home location.
What other possible causes are there?
Battery life?

As for the cause of the incident, I have to wonder if GPS jamming might have been involved.
When the GPS trouble started the drone showed 30 sats but no confidence in the position data.
Then for 7 seconds it showed zero sats, but for the rest of the flight it reported 27-29 sats but no confidence in the position data and did not calculate any position data.

I find it very strange that it later lost power because that is unrelated to false GPS signal.
It's not impossible to have two unrelated rare incidents affect the same flight, but it would be very, very unusual.

After losing GPS 150 metres from home, the drone heading was mostly pointing 70-80°, generally towards the homepoint.
Joystick input was only for 36 seconds (out of nearly 3 minutes), mostly forward flight at medium speeds.
That should have brought the drone closer to the home point and possibly past it, but it's not possible to tell where to.

If I was searching for it, I'd start looking about 200 metres east of the home point and work back to the west.
Best guess at an approximate search area is this triangle.

Mamelo Posted at 10-4 16:15
Some kind of deliberate interference with the signal?
I was flying near a cultural site although not blocked on maps.

GPS jamming is rarely seen, but that's what best fits the unusual data.
It's unlikely to be related to a cultural site.

Labroides Posted at 10-4 16:32
GPS jamming is rarely seen, but that's what best fits the unusual data.
It's unlikely to be related to a cultural site.

So I should look for it on the opposite side of the last known gps coordinates?
It will be near impossible to find it in those woods.

Anyway, thank you very much for your time and help!

Mamelo Posted at 10-4 16:49
So I should look for it on the opposite side of the last known gps coordinates?

That's my best estimate, looking at the joystick inputs and compass data.

Mamelo Posted at 10-4 16:49
So I should look for it on the opposite side of the last known gps coordinates?
It will be near impossible to find it in those woods.

Perhaps JJB might be able to offer some ideas when/if he sees this thread.

One last thing, if I remember correctly while panicking, I think the drone was above the home point and heading towards your search area when the controller lost signal.

Hello there Mamelo. I am sorry to read and to know what happened to your DJI Mini 3 Pro and thank you for reaching out. Since this unfortunate incident happened and you have lost your DJI Mini 3 Pro. It would recommend for you to contact our DJI Support Team at https://www.dji.com/support?site=brandsite&from=nav for further assistance. We have a team that will do there best to help you and give out the best resolution for the said issue. Again, I am sorry for your loss and thank you.

Labroides Posted at 10-4 15:29
I don't enjoy the tough ones like this.
It's a real puzzle with too many pieces missing.

You have a particular set of skills.  And if it was easy, I suspect it would not challenge you.

I enjoy seeing the analysis come together, and appreciate there are people like you up the challenge!

Labroides Posted at 10-4 16:58
Perhaps JJB might be able to offer some ideas when/if he sees this thread.

Hi Labroides,

seeing this log now.

Indeed GPS 0-5 value suddenly to 0, GPS # sats from 30 to zero, but after 7 secs in steps back to 27.But the 0-5 remains at zero. hmm, why? software issue like fixed on zero?
With GPS jamming the GPS lat lon data shows weird data, in this log all the GPS LatLon has the same value.
If this is the same as done to other PhantomHelp converting fields than this is bc data is copy/paste from the last good record into the next records.

About speed in the CSV ; the X Y speed is in the log even with No GPS.
So kind of possible to make an calculated guess distance drone moved after the 18m10.

Around 16m  flying high with no RC input : steady hovering with low drone pitch roll angles = almost no wind.

Question is what happend after the last record?
Drone not in ATTI mode, seen by braking actions releasing sticks.

After disconnected it should RTH, wich it will not as GPS isn`t used.
Think drone will hover and hover, finally land due to low battery.

Will make a distance over time + heading calculation.

Approx 135 meter travel due to RC inputs in the average heading 079.My estimated guess is end of red line is where drone did hover until low batt landing.
But if so....it is really close to Homepoint....
[ DronePosition = last position in N mode ]

OP wrote in post #21 that drone was almost above HP, inline with my calculation.
As it did not land there, why and where did it moved to....

cheers
JJB

So guys, could you tell me which side from to home point to look for it?
Somewhere left of the homepoint where the last coordinates are or right in this triangle area?

Sean-bumble-bee Posted at 10-4 14:28
This is possibly going to be an interesting thread, you might like to post it over on MavicPilots too.
Question for the likes off JJB and Labroides, after 18:12 can the indicated speeds be relied upon?

Hi,

In the CSV 2 columns ( OSD.xSpeed [MPH], OSD.ySpeed [MPH]) about speed, in the X and Y direction of movement. Result of the 2 values is the "XY" speed.

FRAP calculates GPS speed (distance over time between a record to the next record)

See my chart, XY and GPS speed 2 lines. Until GPS to zero line on top of each other, meaning that GPS caluclated speed = XY calculated speed derived from data in the 2 values.

cheers
JJB

Mamelo Posted at 10-5 01:15
So guys, could you tell me which side from to home point to look for it?
Somewhere left of the homepoint where the last coordinates are or right in this triangle area?

Hi

In my chart is the DronePostion = last postion in N-mode, after this you move the drone for about 135 meter over the red line.

So just before disconnect your drone was almost above HomePoint. (end of the red line)

What happed after i do not know. In my logic it did hover there al the time.
If the last wind warning is correct than it could drift away from position if the drone could not handle the wind.
Too little data in the log to calculate strong wind drift correction.
cheers
JJB

JJB* Posted at 10-5 00:30
Hi Labroides,

seeing this log now.

Question is what happend after the last record?
.....
After disconnected it should RTH, wich it will not as GPS isn`t used.
But why was signal lost with the drone so close and signal never regained?

Labroides Posted at 10-5 02:53
Question is what happend after the last record?
.....
After disconnected it should RTH, wich it will not as GPS isn`t used.

good question....

guess we will never find that answer!

Mayby DJI knows more....

He was only about 55m high at the end of the log, perhaps with the low height, the trees killed LOS,maybe?

I'm not saying that this had anything to do with the case at hand, but the Turks have developed a system of protecting military and cultural sites from drone overflight.  Here's a video of the system in action:


Since this incident occurred near a cultural site, the system should be considered.  Here's a video of the system in use near a cultural site in Istanbul. Go to about 6:50 to see the incident.


We just need to be aware that such systems exist and the incident in this thread as it was near a cultural center could have been related to this tech.

Wow this is interesting, although I highly doubt our authorities have these devices

Bashy Posted at 10-5 07:32
He was only about 55m high at the end of the log, perhaps with the low height, the trees killed LOS,maybe?

He was standing in a large open area,on the highest point around, so this is not what happened.

Mamelo Posted at 10-5 09:20
Wow this is interesting, although I highly doubt our authorities have these devices

Hi there neighbor! Back in the day, I didn't look at the markets in Croatia, but in Serbia and Bosnia, it was relatively easy to purchase GPS jammers until maybe 3-4 years ago. Perhaps someone illegally obtained one and used it?

About the GPS behavior, with position showing...
The drone has receiver configured to receive GPS, Galileo and BeiDou.
What would happen if one of those systems started reporting position which was much different from the other two? How would the drone behave? I don't remember seeing any documentation from DJI on that.

For authorities, well, if they got the drone, they could come into contact with you, using the drone operator ID number which you should have on the drone.

Labroides Posted at 10-5 14:08
He was standing in a large open area,on the highest point around, so this is not what happened.

Oops, my bad, i should have looked on Google Earth 1st like i normally do, didn't realise it was quite a bit higher than the surrounding trees.

Mamelo Posted at 10-5 01:15
So guys, could you tell me which side from to home point to look for it?
Somewhere left of the homepoint where the last coordinates are or right in this triangle area?

Hi,

Had a second thought...

What if your drone did start a RTH using data just before the GPS went off.
RTH started almost above HP, using distance/bearing  it will fly to the marker on my map.
Cannot explain it in simple words, but mayby a try to do a search there

cheers
JJB

JJB* Posted at 10-5 23:33
Hi,

Had a second thought...
[Image]

I searched there yesterday but no luck because of a difficult terrain. I'll give it a try again.

Thank you!

Mamelo Posted at 10-6 01:07
I searched there yesterday but no luck because of a difficult terrain. I'll give it a try again.

Thank you!

Just out of interest, do you have Care Refresh with Flyaway cover?

Bashy Posted at 10-6 03:31
Just out of interest, do you have Care Refresh with Flyaway cover?

Well I bought it with my drone in the store, but they didn't send me the activation code and I didn't activate it before the flight. So I think I can't ask for a Flyaway coverage...I wouldn't even search for it if I was eligible for a replacement

JJB* Posted at 10-5 23:33
Hi,

Had a second thought...

What if your drone did start a RTH using data just before the GPS went off.
That sounds unlikely.
1.  GPS would have to returned to normal.
2.  Why was signal lost?
3.  Wouldn't the flyer have heard/seen it if it did RTH?