Nice calm sunday evening. I started to gain altitude directly upwards. Somewhere in 40m height my drone lost control. Just like I have read before. No control at all. I'm not sure if the video link was working or not. Fortunately the drone stopped on 200m distance. I Walked toward the drone which was behind the trees. Then I suddenly got the control back. There is an oil refinery 1,5km away. I'm glad it didn't go there.

LOGS:

https://www.phantomhelp.com/logviewer/D7DEA1LZ8J639Q5QUGUI/

PS. How can I select "Mavic Mini" Topic type? Looks like you can't if using a computer. With a mobile phone I had more options. I'm sorry for posting to wrong topic.

Nice calm saturday evening. I started to gain altitude directly upwards. Somewhere in 40m height my drone lost control. Just like I have read before. No control at all. I'm not sure if the video link was working or not. Fortunately the drone stopped on 200m distance. I Walked toward the drone which was behind the trees. Then I suddenly got the control back. There is an oil refinery 1,5km away. I'm glad it didn't go there.

LOGS:

https://www.phantomhelp.com/logviewer/D7DEA1LZ8J639Q5QUGUI/

Not sure what happened, can’t see signal drop, speed shows close to 43mph but if you look at how far drone was traveling it was not actually going 43mph so telemetry was giving wrong reading. Labroids might come along and have a closer look. Gps signal looks ok but if your RC line of sight to craft was blocked by trees this will cause loss of signal or video or both.

hallmark007 Posted at 8-3 04:11
Not sure what happened, can’t see signal drop, speed shows close to 43mph but if you look at how far drone was traveling it was not actually going 43mph so telemetry was giving wrong reading. Labroids might come along and have a closer look. Gps signal looks ok but if your RC line of sight to craft was blocked by trees this will cause loss of signal or video or both.

I checked the log file. I noticed that SMART_BATTERY.usefulTime  changed from 0 to 1371 at the same time when it ran away.

You received many noticed of  Aircraft in high interference environment. Fly with caution(Code: 80015). Aircraft signal interference,  Weak signal. Adjust antennas(Code: 80016).

Clearly that was your problem. In those cases make sure you set your Signal disconnection option for the most appropriate action (default RTH) and ALWAYS and AT ALL TIMES determine and input into the Fly app the RTH height for your intended flight environment and path.

The Mini ascended above the home point and as it passed 34 metres it bagan to move away from the home point.
At 35 metres the Mini is moving at 0.5 m/s
At 37.8 metres it is moving at 1.7 m/s
At 38.5 metres it is moving at 2.9 m/s
This is a significant increase in speed with only 3.5 metres altitude increase.
It might be explained by a gusting wind ...  except that the speed increases rapidly.
At 39.6 metres the Mini has reached the normal Max Speed for P Mode (8 m/s) and continues to increase as the drone is brought down lower.

At 38.2 metres (0:27.2) the mini has reached normal Max Speed for Sport Mode (13 m/s) and continues to increase to 19.3 m/s at 0:29.7.
Oddly this happens with very gently pitch angles, much less than is normally needed for even normal speeds.

By 0:50.3 the incident has finished and the crazy speed has stopped and the drone hovers and responds to controls normally for the rest of the flight.

The incident puzzles me.
It has some elements of a normal yaw error situation but significant differences.

I have two questions:
1.  What was the launch surface, what steel objects might have been within a metre of the launch point?
2.  What direction was the drone pointing while on the ground at launch?
Was it NE or some other direction?

I'd suggest synching your flight data and opening a case with DJI to have their experts loo into it.

I've noticed quite a few reports of connection issues with mini's when flying over boats or harbours like the one in your log.

I'd guess there is some issue with high power nautical transmitters

m80116 Posted at 8-3 05:30
You received many noticed of  Aircraft in high interference environment. Fly with caution(Code: 80015). Aircraft signal interference,  Weak signal. Adjust antennas(Code: 80016).

Clearly that was your problem. In those cases make sure you set your Signal disconnection option for the most appropriate action (default RTH) and ALWAYS and AT ALL TIMES determine and input into the Fly app the RTH height for your intended flight environment and path.

You received many noticed of  Aircraft in high interference environment. Fly with caution(Code: 80015). Aircraft signal interference,  Weak signal. Adjust antennas(Code: 80016).

Clearly that was your problem

This member specialises in making crazy statements because he has a very incomplete understanding of how drones work and an exaggerated idea of his level of expertise.
Ignore him and his suggestion.
Interference doesn't take control of your drone and fly away with it.
Interference, at worst will block out your control signal which will simply initiate RTH.

itchyeyeballs Posted at 8-3 06:09
I've noticed quite a few reports of connection issues with mini's when flying over boats or harbours like the one in your log.

I'd guess there is some issue with high power nautical transmitters

I've noticed quite a few reports of connection issues with mini's when flying over boats or harbours
The Mini has a weak wifi controller signal.
It is subject to connection issues anywhere, not just near boats.

itchyeyeballs Posted at 8-3 06:09
I've noticed quite a few reports of connection issues with mini's when flying over boats or harbours like the one in your log.

I'd guess there is some issue with high power nautical transmitters

It’s very unusual to pick up interference when craft is in the air, but very common to pick up interference on the ground and not show until your craft well into it’s flying mission.

You may see interference for example if you tried to land on a steel roof or fly very close to electric pylons. But it’s very unusual.

I launched it from the wooden roof of my boat.  There is a light mast made of stainless steel on the roof. Maybe 1 meter away.  The weather was very calm and no gusting winds. The drone was pointing to NE if I remember right. My mobile phone was in flight mode to increase the signal quality. I have 30m RTH height and return to home point. There was no signal interference before and after the run away. During the incident I tried to steer it to all directions. I don't know if it's visible on the log file?

I'm not so stupid that I think it's normal in a error situation that the drone will fly away as fast as possible. This is a place where one of the biggest cruise ships are built.

Labroides Posted at 8-3 06:13
I've noticed quite a few reports of connection issues with mini's when flying over boats or harbours
The Mini has a weak wifi controller signal.
It is subject to connection issues anywhere, not just near boats.

Agreed but there have been several examples where users report much worse performance when flying in those sort of areas.

I doubt it directly caused the strange issues reported in this thread though but may have contributed to the signal loss.

Could they have anti drone devices operating at that shipyard? must be commercially sensitive?

get too close and signal jammed?

itchyeyeballs Posted at 8-3 07:06
Could they have anti drone devices operating at that shipyard? must be commercially sensitive?

get too close and signal jammed?

There are two thousand people from 30 different countries working on the dockyard. I don't think it makes sense to jam drones. This is not marked as a "no drone zone".  And if they would mess with the GPS signal there will be one thousand angry boaters trying to navigate out from the marina nearby.

itchyeyeballs Posted at 8-3 07:06
Could they have anti drone devices operating at that shipyard? must be commercially sensitive?

get too close and signal jammed?

Could they have anti drone devices operating at that shipyard? must be commercially sensitive?
get too close and signal jammed?
That would be about as likely as being caught by the tractor beam from an alien spacecraft.

hallmark007 Posted at 8-3 06:16
It’s very unusual to pick up interference when craft is in the air, but very common to pick up interference on the ground and not show until your craft well into it’s flying mission.

You may see interference for example if you tried to land on a steel roof or fly very close to electric pylons. But it’s very unusual.

It’s very unusual to pick up interference when craft is in the air, but very common to pick up interference on the ground and not show until your craft well into it’s flying mission.
I suspect you are confusing signal interference with magnetic interference.
Signal interference is often detected in the air and is common in urban and industrial areas.

JussiSL Posted at 8-3 06:46
I launched it from the wooden roof of my boat.  There is a light mast made of stainless steel on the roof. Maybe 1 meter away.  The weather was very calm and no gusting winds. The drone was pointing to NE if I remember right. My mobile phone was in flight mode to increase the signal quality. I have 30m RTH height and return to home point. There was no signal interference before and after the run away. During the incident I tried to steer it to all directions. I don't know if it's visible on the log file?

I'm not so stupid that I think it's normal in a error situation that the drone will fly away as fast as possible. This is a place where one of the biggest cruise ships are built.

That looks like a very interesting place to fly.

I launched it from the wooden roof of my boat.  There is a light mast made of stainless steel on the roof. Maybe 1 meter away.
The drone was pointing to NE if I remember right.
That shouldn't have caused a problem.
I was checking for things like launching from reinforced concrete or on a wooden surface with plain steel bolts close to the compass of the drone.
I don't think this was a yaw error incident but was just checking to help eliminate that.

During the incident I tried to steer it to all directions. I don't know if it's visible on the log file?
It was very obvious.

Labroides Posted at 8-3 08:01
It’s very unusual to pick up interference when craft is in the air, but very common to pick up interference on the ground and not show until your craft well into it’s flying mission.
I suspect you are confusing signal interference with magnetic interference.
Signal interference is often detected in the air and is common in urban and industrial areas.

I was talking about magnetic interference, someone had mentioned yaw problems.

@Labroides

Not sure if it's part of your normal viewing scope (or something that's only in the dat's) but was the Drone overwhelmed / lagging at all during the acceleration period?

Labroides Posted
This member specialises in making crazy statements because he has a very  incomplete understanding of how drones work and an exaggerated idea of  his level of expertise.
Ignore him and his suggestion.
Interference doesn't take control of your drone and fly away with it.
Interference, at worst will block out your control signal which will simply initiate RTH.

You're retarded and having cognitive problems. After statement you said the exact same thing.

Quote:
The Mini has a weak wifi controller signal.
It is subject to connection issues anywhere, not just near boats.

By the way your mind is moving at about 30 m/s inside your brain box. I suggest you a complete overhaul of your cranium as it's either A) not the right size for your one-celled brain or B) completely vacant of any suitable matter.

That's how we deal with idiots like you baby.

Gentlemen, check for yourself the history of these users: Labroides, hallmark007. You'll surely find many posts alwasy offending first, always trolling, always personal attacking. DJI moderation NEVER took a stance against them.

Labroides Posted at 8-3 06:11
You received many noticed of  Aircraft in high interference environment. Fly with caution(Code: 80015). Aircraft signal interference,  Weak signal. Adjust antennas(Code: 80016).

Clearly that was your problem

You like to make a complete FOOL of yourself making false claims that can't stand the most basic of proof.

Nobody in their right mind would say this is normal flight log with nothing wrong about signal interference.

JussiSL Posted at 8-3 07:35
There are two thousand people from 30 different countries working on the dockyard. I don't think it makes sense to jam drones. This is not marked as a "no drone zone".  And if they would mess with the GPS signal there will be one thousand angry boaters trying to navigate out from the marina nearby.

scratch that theory then

m80116 Posted at 8-3 11:12
You like to make a complete FOOL of yourself making false claims that can't stand the most basic of proof.

Nobody in their right mind would say this is normal flight log with nothing wrong about signal interference.

Is any of these error messages GPS related? My home point was updated ok. Why my drone didn't came back to the home point after the signal loss? I think this looks like a software bug. Other users have also reported that their Minis just flew away. Someones reached 60 mph. How could I know before hand that there is interference? It's very much useless information to know that now there is interference if the drone flies away immediately.

Labroides Posted at 8-3 06:11
You received many noticed of  Aircraft in high interference environment. Fly with caution(Code: 80015). Aircraft signal interference,  Weak signal. Adjust antennas(Code: 80016).

Clearly that was your problem

"This member specialises in making crazy statements because  he has a very  incomplete understanding of how drones work and an  exaggerated idea of  his level of expertise.
  Ignore him and his suggestion."

Well, I do agree that the warning messages were not very important here because the first one did not occur until the issue had passed. So there was no advance warning to heed.

However, you could have put that a bit more delicately instead of treading right over him like that. Not cool.

JussiSL Posted at 8-3 12:10
Is any of these error messages GPS related? My home point was updated ok. Why my drone didn't came back to the home point after the signal loss? I think this looks like a software bug. Other users have also reported that their Minis just flew away. Someones reached 60 mph. How could I know before hand that there is interference? It's very much useless information to know that now there is interference if the drone flies away immediately.

Is any of these error messages GPS related?
None are and none would be expected to have much impact on a drone flight.
They are all the kind of thing you would expect flying around a big shipyard.

My home point was updated ok. Why my drone didn't came back to the home point after the signal loss?
I can't find any time where you lost signal.
It was connected for all of the flight.

I think this looks like a software bug.
I'm not certain of that but it's strange enough that I suggested you refer it to DJI's analysis people.
If it was a software bug, we'd see similar incidents from others (but we don't).

Other users have also reported that their Minis just flew away. Someones reached 60 mph. How could I know before hand that there is interference? It's very much useless information to know that now there is interference if the drone flies away immediately.
To avoid a yaw error, don't launch from steel or reinforced concrete surfaces.
You can confirm things are good for launch by checking that the drone is facing the same way as the drone icon on the map before launching (eg .. drone faces east .. drone icon faces east)

Your incident had some similarity to a yaw error incident but was also quite different.
I've analysed hundreds of flight incidents but never seen one like this.

GaryDoug Posted at 8-3 12:48
"This member specialises in making crazy statements because  he has a very  incomplete understanding of how drones work and an  exaggerated idea of  his level of expertise.
  Ignore him and his suggestion."

However, you could have put that a bit more delicately instead of treading right over him like that. Not cool.
That particular member continues to give really bad "advice" and goes ballistic if anyone contradicts it with facts.
He continues to do it and gets worse over time.
I have been delicate with him and you can see where that's got me.
What I stated was accurate.

GaryDoug Posted at 8-3 12:48
"This member specialises in making crazy statements because  he has a very  incomplete understanding of how drones work and an  exaggerated idea of  his level of expertise.
  Ignore him and his suggestion."

Well you can surely see that your friend seems to have no problem bringing all and sundry into his argument. I came to this thread to do nothing but give op some advice and try answer some of his questions.
But you friend then out of nowhere decides to bring me into his argument insult me into the bargain and why. Well I’ll tell you why, because he has lackeys like you who continually accept everything he says completely ignore the fact that he spends much of his time on this forum abusing others saying what he likes out the side of his mouth.
He then further to that accuses almost everyone of being the author of their own downfall, this from the same guy who spent months on this forum accusing dji for the crash of his drone, looking for members support but refusing to show any flight log because it showed he was at fault. He was then given good advice but chose to insult members throw their help back in their face, quoting “ you guys are to old to tell me what to do”
He’s a complete idiot and he does not care about you or anyone else around here a complete narcissist with the personality of a bean, and if he’s getting it hot it’s because he spends much of his time here not helping anyone but going around like the  cat who got the cream.

I think you have approached this completely blinkered and that’s a pity, because if you had bothered to read what your friend wrote about others you might have seen him for the complete narcissistic fool he really is .

As far as I can see, the problems began when the drone got conflicting velocity calculations from the IMU and the GPS. At least that is how I interpret the fields

gpsVelN-velN   and  gpsVelE-velE

The differences here are in the range of up to +/- 5 m/s and more (!). I would guess that N and E here stands for North and East. The IMU itself (by integration of the accelerometer readings) can only directly measure the velocities in the x and y axis of the "airframe", it needs at least an initial compass reading to translate this to velocities in a North and South coordinate system.
So is the compass to blame?
Perhaps, but note that all this happens while the number of received GPS sats jumps between 8,9,and at most 10.
The drone seems to be using GPS all the time, so I wonder if people here who are more familiar with these things will want to look into the hypothesis that this was caused by the drone "overreacting" to poor GPS data.



If true, it would still be a "bug" IMHO, so opening a ticket for DJI to look into this might be a good thing to do. Maybe there are ways to make the drone more fail-safe in the event of bad GPS reception while in flight?

Fly safe
HB

Wifi connection is nothing like the occusync connection.  Wifi you have to have a direct LOS between RC and Aircraft, meaning if you were behind a tree that is a big no no for wifi, you would have been fine if it was an occusync connetion.

hallmark007 Posted at 8-3 16:08
Well you can surely see that your friend seems to have no problem bringing all and sundry into his argument. I came to this thread to do nothing but give op some advice and try answer some of his questions.
But you friend then out of nowhere decides to bring me into his argument insult me into the bargain and why. Well I’ll tell you why, because he has lackeys like you who continually accept everything he says completely ignore the fact that he spends much of his time on this forum abusing others saying what he likes out the side of his mouth.
He then further to that accuses almost everyone of being the author of their own downfall, this from the same guy who spent months on this forum accusing dji for the crash of his drone, looking for members support but refusing to show any flight log because it showed he was at fault. He was then given good advice but chose to insult members throw their help back in their face, quoting “ you guys are to old to tell me what to do”

"....he has lackeys like you..."

Who are you addressing? And what is this friend thing about? Are you saying I am his friend? I am not but that is not the issue. I am a member of this forum and have a right to complain about abusive behavior like yours and one other. It is obvious that there are just 2 members who behave this way. Now you choose to insult me just because it is your nature to be rude I guess. That pretty much proves the point doesn't it. Rude behavior must be ok with your type but it is not for everyone. Go back to sleep dude.

Hello there JussiSL. I am sorry for the trouble this has caused and thank you for reaching out. It is great to know that you were able to recover your DJI Mavic Mini with no unwanted incident. Can you please try flying your DJI Mavic Mini to a different location with less interference to see if the issue will still persist? The said DJI drone might have picked up magnetic interference from the area. That might be the reason why you are experiencing these issues. In addition, I will post an official DJI tutorial video on DJI Remote Controller Antenna Best Practices. Thank you and fly safe always.

DJI Stephen Posted at 8-3 22:48
Hello there JussiSL. I am sorry for the trouble this has caused and thank you for reaching out. It is great to know that you were able to recover your DJI Mavic Mini with no unwanted incident. Can you please try flying your DJI Mavic Mini to a different location with less interference to see if the issue will still persist? The said DJI drone might have picked up magnetic interference from the area. That might be the reason why you are experiencing these issues. In addition, I will post an official DJI tutorial video on DJI Remote Controller Antenna Best Practices. Thank you and fly safe always.

https://www.youtube.com/watch?v=P7M9JtfVwQE

In addition, I will post an official DJI tutorial video on DJI Remote Controller Antenna Best Practices.
Why would you post that?
Proper antenna orientation is always important.
But it's completely irrelevant to this incident.

GaryDoug Posted at 8-3 18:39
"....he has lackeys like you..."

Who are you addressing? And what is this friend thing about? Are you saying I am his friend? I am not but that is not the issue. I am a member of this forum and have a right to complain about abusive behavior like yours and one other. It is obvious that there are just 2 members who behave this way. Now you choose to insult me just because it is your nature to be rude I guess. That pretty much proves the point doesn't it. Rude behavior must be ok with your type but it is not for everyone. Go back to sleep dude.

Again you ignorance shows through and your self enlightenment is clear. You ignored reading the thread and decided to voice your opinion on whatever suited your ridiculous argument. You ignored that your friend was both insulting and baiting others without any provocation and why, because you feel you have some authority.
I can tell you your insistence on separating and commenting on only things that suit your own argument shows nothing but a real ignorance on your part. Now I suggest you go back and read the thread take the blinkers off and comment in an impartial way, not in a way you think makes you favour of the month. You want to be forum police do it correctly or be prepared to show how much of a complete ignoramus you are.

I suggest you go back to post 20 read it and try figure out why my name was brought into this, it seems you don’t like this behavior so comment on it, you seem to comment willingly when it suits you.

JussiSL Posted at 8-3 12:10
Is any of these error messages GPS related? My home point was updated ok. Why my drone didn't came back to the home point after the signal loss? I think this looks like a software bug. Other users have also reported that their Minis just flew away. Someones reached 60 mph. How could I know before hand that there is interference? It's very much useless information to know that now there is interference if the drone flies away immediately.

Let me address your concerns.

- The messages you received were not GPS related. The GPS reception was actually good to optimal throughout the flight.

-  You drone didn't come back to you because you never disconnected from  it completely (if you've uploaded the right flight log). Anyway even  after a complete disconnect it takes about 11 seconds before the drone  initiates the RTH heading alignment, climb to preset height and begins  flying towards the home point.

- You can easily check yourself the home-point environment from your Mobile Device, just go to the Settings (three dots menu) >> Transmission (tab)  here you can see the frequency that the Mini auto-selected at power-up  (only. dynamic swtiching is NOT supported) and see which channels have  the most activity, signaled by a red bar. Observe for couple of minutes,  of all the bars rise and become red alternatively it means you're  chances to experience a smooth and distant flight will be quite low. A  bar oscillating and alternating between red and green it's not an  indication of a partly good channel, rather that channel experiences  pulses of modulation that are likely to disrupt your signal quality even  more than one where it stays red (where at least the reach limit will  be much closer but stable).

- The drone DOES not fly away for  signal interference or weak signal, but you must make sure you set your  RTH height according to your environment, taking into account water  towers, electricity and telephone poles and any other obstacle that it  might encounter. Also it's very useful to make sure your Signal Lost  option is set for RTH (Return To Home) and not to hover or land where it  is.

I just realized that csvView has a nice way to visualize this effect in the GEOplayer.

The tool allows you to display the flightpath based on a) GPS readings (here red) and b) just based on the IMU velocity calculations (here green).
As you can see, both diverged quickly (about 23 seconds into the flight, when the number of GPS sats dropped back to 8, see plot posted earlier).
When the nr of sats finally rose  to 14, IMU and GPS were in agreement again about the drone's velocity (if you shifted the IMU and GPS flighpaths after this point, they would match again).

For comparison, this is the same plot for one of my uneventful flights (at roughly the same map scale)

The WiFi transmission was never a problem. In fact the drone kept contact with the RC and tried to follow the pilot's slightly panicy stick inputs (it pretty much follows up/down commands) but  at the same time the drone tries tio react to wrong assumptions about its own horizontal velocity.

m80116 Posted at 8-4 03:34
Let me address your concerns.

- The messages you received were not GPS related. The GPS reception was actually good to optimal throughout the flight.

Yes, I have read the manual and I understand how it works. But I think that this has nothing to do with the problem I had. The drone is not supposed to fly on its own to somewhere. Do you have any thoughts about the incident?

EHEA Posted at 8-4 03:46
I just realized that csvView has a nice way to visualize this effect in the GEOplayer.

The tool allows you to display the flightpath based on a) GPS readings (here red) and b) just based on the IMU velocity calculations (here green).

Thank you for the very nice analysis. I can't see it from the picture but both IMU and GPS were at the same position at the start? Then somehow the IMU position jumped about 30 meters to NW. If the flight controller program was more clever it would understand that it's not possible to move so fast. A full manual control could be a nice feature. I have another drone with Naze32 controller. It has no GPS and it doesn't calculate the speed. It just levels the drone. It doesn't drift very much in calm weather.
Edit: I took a look at the picture again. It looks like the IMU position doesn't jump. It goes smoothly to wrong direction. Maybe the compass was pointing to a wrong direction? Usually it asks to make calibration but not this time.

JussiSL Posted at 8-4 04:12
Thank you for the very nice analysis. I can't see it from the picture but both IMU and GPS were at the same position at the start? Then somehow the IMU position jumped about 30 meters to NW. If the flight controller program was more clever it would understand that it's not possible to move so fast. A full manual control could be a nice feature. I have another drone with Naze32 controller. It has no GPS and it doesn't calculate the speed. It just levels the drone. It doesn't drift very much in calm weather.

Yes at the start, the positions match, but that is by definition. The IMU can only sense acceleration and rotation (and velocity from doing math on the former values), not location, so the initial starting point in the plots is taken from GPS.

The divergence between IMU and GPS does not happen in one big jump, but rather gradually while  the drone is flying away, that is shown in the graph I posted earlier about the differences of the  gpsVelocity and the IMUvelocity. This coincides with the period when the drone got only 8 or 9, sometimes 10 sats, and vanishes when the drone receives > 11 sats, so I rather suspect that the somewhat inaccurate GPS values were bouncing around quite a bit during this incident, which led the drone to "think" it was moving at crazy speeds for short momenets (up to 43 mph!) while actually it wasn't.

From an engineering point of view, this can be a rather challenging situation with a lot of coupling and bad feedback in the control loop.
Think about it: the GPS positions bounce around so the drone thinks there is a high velocity it needs to counter-act. The result is that the drone is tilting a lot to correct for this. The tilt will change the orientation of the GPS antenna ... oh .oh... this can lead to a different set of satellites being used when there are already few available ==> another jump in the (inaccurate locations) ==> the drone again thinks it needs to correct for sone spurious velocity change ==> changes the tilt, changes the GPS antenna orientation .... if this is not dampened correctly, it's easy to believe that this can lead to "surprising" drone movements.

JussiSL Posted at 8-4 03:56
Yes, I have read the manual and I understand how it works. But I think that this has nothing to do with the problem I had. The drone is not supposed to fly on its own to somewhere. Do you have any thoughts about the incident?

Yes, kind of... what I've tried to tell you is that I can't find any trace of complete disconnect in your flight log. Also channels and frequency selection operation is not described in the manual, we had to discover that ourselves.

I also cannot see any clear sign of erratic behavior, just slight traces but it could just be inertia and wind interaction (about 5 to 6 m/s at the time). But it could also be (and I mean beginning at 40sec. of flight) the cause of powering the drone in a distorted magnetic field.

I notice your Mini took off from a pier... if you also turned it on above or anywhere near it, it could well be the magnetic field reference which was wrong. It's something that I personally wouldn't have done, as I don't like risking my chances, not even after verifying the magnetic heading with my MD. Ideally you should have powered on your Mini on the road (at your height) and walk it to your starting point.

EHEA Posted at 8-4 04:36
Yes at the start, the positions match, but that is by definition. The IMU can only sense acceleration and rotation (and velocity from doing math on the former values), not location, so the initial starting point in the plots is taken from GPS.

The divergence between IMU and GPS does not happen in one big jump, but rather gradually while  the drone is flying away, that is shown in the graph I posted earlier about the differences of the  gpsVelocity and the IMUvelocity. This coincides with the period when the drone got only 8 or 9, sometimes 10 sats, and vanishes when the drone receives > 11 sats, so I rather suspect that the somewhat inaccurate GPS values were bouncing around quite a bit during this incident, which led the drone to "think" it was moving at crazy speeds for short momenets (up to 43 mph!) while actually it wasn't.

Next time I will wait for more satellites. Usually GPS doesn't get the speed from the position data. It's calculated from the Doppler shift of the signal frequency. Maybe it's not possible to get this accurate speed information for all three axes. I guess that DJI doesn't make own GPS receivers.