So I had just lifted off, ready to go south to fly over a plant that makes gravel. No wind at all. But just as I get to the height I want, the Mini inexplicably takes off to the west, nearly hitting a railroad retaining wall. As quickly as I could, I was trying to raise the height and did so right before the wall. On the remote screen it said 'radio interference'. After I brought it back to me, I then kept flying (cautiously) and didn't have other issues that session. Would interference do this? Or is the RC faulty?

Would interference do this? Or is the RC faulty?
Interference won't ever cause your drone to fly off uncontrolled and it's unlikely to be your controller.
Flight data is needed to confirm but the cause is probably a yaw error caused by magnetic distortion at your launch point.

Post your flight data and describe the spot you launched from if you want to find out more.

There could be something wrong with the drone. Is this the first time is has happened to you?

Definitely the flight data would help.  Three things I would definitely check are if your RC firmware, Mini firmware and DJI Fly app are update.  There was a rash of crashes that coincided with the DJI Fly app updates 1.2.0 and also with the RC firmware update for the Air 2/Mini 2.  

I had the same weird takeoff similar to what you are showing, but less aggessive, after the RC update on 11/5 and it was fixed with the update on 11/7.  I did the same thing and brought it back and landed.  Then calibrated everything and it did it again.  After I got up in the air flew without issue albeit a short distance before other weird flight telemetry caused me to initiate an RTH.  In the air the issue were not sever but the Air 2 was behaving slightly off and laggy.

We may not have had the same issue as they are different drones but the same RC.

Labroides Posted at 11-19 17:19
Would interference do this? Or is the RC faulty?
Interference won't ever cause your drone to fly off uncontrolled and it's unlikely to be your controller.
Flight data is needed to confirm but the cause is probably a yaw error caused by magnetic distortion at your launch point.

Ok, thanks for the reply. I'll take a look at the Flight Data now.

JohnLietzke Posted at 11-19 17:37
Definitely the flight data would help.  Three things I would definitely check are if your RC firmware, Mini firmware and DJI Fly app are update.  There was a rash of crashes that coincided with the DJI Fly app updates 1.2.0 and also with the RC firmware update for the Air 2/Mini 2.  

I had the same weird takeoff similar to what you are showing, but less aggessive, after the RC update on 11/5 and it was fixed with the update on 11/7.  I did the same thing and brought it back and landed.  Then calibrated everything and it did it again.  After I got up in the air flew without issue albeit a short distance before other weird flight telemetry caused me to initiate an RTH.  In the air the issue were not sever but the Air 2 was behaving slightly off and laggy.

It's funny, because I've been purposefully avoiding updating because of the issues after install. It's good to hear you say that the most recent update seems to have solved some issues.

djiuser_D0sgAm6xL5jn Posted at 11-19 17:24
There could be something wrong with the drone. Is this the first time is has happened to you?

Yep. First time. I did question myself and wonder if I wasn't paying attention to my thumbs on the joystick, but that certianly wasn't the case.

DGCA3 Posted at 11-19 19:04
Yep. First time. I did question myself and wonder if I wasn't paying attention to my thumbs on the joystick, but that certianly wasn't the case.

Yeah, you might want to check in with DJI about this.

DGCA3 Posted at 11-19 18:56
Ok, thanks for the reply. I'll take a look at the Flight Data now.

To get to your flight data, go to: https://www.phantomhelp.com/LogViewer/Upload/
... where you'll find instructions to upload the data to that site and it will give a report
For help interpreting the data you can post a link for the report here.
Or just post the .txt file.

And please describe the spot where you launched as that's important.
It's much more likely to be the cause rather than the suggestions regarding firmware that you've been offered.

DGCA3 Posted at 11-19 18:59
It's funny, because I've been purposefully avoiding updating because of the issues after install. It's good to hear you say that the most recent update seems to have solved some issues.

The RC firmware was causing the drone to go bonkers and it would even move while hovering 15 to 20 feet then correct itself.  At the the time of the RC firmware update the .360 Mavic Air 2 firmware, I think it was, had been on my drone for a couple of months and I never experienced this.  It was like the the RC was sending phantom signals in the background.  

I contacted DJI with in an hour or two of the firmware being released it was so concerning.  But today the weather was nice and I took a longer flight over Apple's Circular Campus and the signal was better and more stable than previous flights and the drone behaved properly.  The problem has been isolated to the RC hence the second update on 11/7.

JohnLietzke Posted at 11-19 19:40
The RC firmware was causing the drone to go bonkers and it would even move while hovering 15 to 20 feet then correct itself.  At the the time of the RC firmware update the .360 Mavic Air 2 firmware, I think it was, had been on my drone for a couple of months and I never experienced this.  It was like the the RC was sending phantom signals in the background.  

I contacted DJI with in an hour or two of the firmware being released it was so concerning.  But today the weather was nice and I took a longer flight over Apple's Circular Campus and the signal was better and more stable than previous flights and the drone behaved properly.  The problem has been isolated to the RC hence the second update on 11/7.

Thanks for the details of your experience.

Did you see any other drones over Apple? I'd imagine that it would be a popular place to drone over.

DGCA3 Posted at 11-19 18:56
Ok, thanks for the reply. I'll take a look at the Flight Data now.

If it is the RC firmware issue the flight logs will only reflect the data that is extractable for the RC controller and it appear normal.  After the incident I reviewed the log and uploaded them to another forum where other members who are far more skilled at interpreting logs came to the same conclusion that nothing appears wrong with the flight.  

The Mavic Air and Mini 2 .dat files that can be downloaded from the drone using the DJI Assistant 2 are encrypted.  Unfortunately, a public decryption key has not been made available to compare the drones data to the RC data.  

Your logs would resolve whether it was an initiated command that cause the drone to erratically takeoff or if it was a hardware issue.

JohnLietzke Posted at 11-19 19:44
If it is the RC firmware issue the flight logs will only reflect the data that is extractable for the RC controller and it appear normal.  After the incident I reviewed the log and uploaded them to another forum where other members who are far more skilled at interpreting logs came to the same conclusion that nothing appears wrong with the flight.  

The Mavic Air and Mini 2 .dat files that can be downloaded from the drone using the DJI Assistant 2 are encrypted.  Unfortunately, a public decryption key has not been made available to compare the drones data to the RC data.  

Ok, thanks. I'm looking at getting the proper specific data. In the meantime, this is just a summary of the flight in Airdata. A max power load and interference are the only notifications that I saw and that were recorded on this summary. On another tab, it shows my drone hitting 45 mph west,  with just a 4 mph southerly wind. That thing was hauling when it went bonkers at the 22 second mark.

I had a similar issue with the Mavic Pro 1 after take off it just went crazy and was flying in all different directions and the only control I had was up and down. I managed to take it quite high about 100 meters and it stabilised itself. when checking my logs it pointed to compass issues because I had taken off so close to many vehicles and a whole of of steel posts around me. Never made that mistake ever again.

Suren Posted at 11-19 20:03
I had a similar issue with the Mavic Pro 1 after take off it just went crazy and was flying in all different directions and the only control I had was up and down. I managed to take it quite high about 100 meters and it stabilised itself. when checking my logs it pointed to compass issues because I had taken off so close to many vehicles and a whole of of steel posts around me. Never made that mistake ever again.

Thanks. Yes, I've heard not to do the calibration around metal. But I was in a cul-de-sac on a undeveloped street with no buildings or street poles right near me. After it goofed, it was ok for the next 5-6 minutes of the flight.

DGCA3 Posted at 11-19 20:08
Thanks. Yes, I've heard not to do the calibration around metal. But I was in a cul-de-sac on a undeveloped street with no buildings or street poles right near me. After it goofed, it was ok for the next 5-6 minutes of the flight.

Hours later and still no data?
Discussing guesses without data that would show what happened is wasting time and effort.
If you'd just post a link to your Airdata report instead of a picture of it, I could probably obtain the data from that.

Labroides Posted at 11-19 20:26
Hours later and still no data?
Discussing guesses without data that would show what happened is wasting time and effort.
If you'd just post a link to your Airdata report instead of a picture of it, I could probably obtain the data from that.

It's the first time I've accessed this stuff. I'm a newbie with data.


https://app.airdata.com/share/uKprcT

Wow that was going as you said "like a bat out of hell" with no right stick. And how does a Mini 2 go 45 mph?.... with a big wind maybe. Something doesn't add up. Lab will probably come up with the best analysis...or guess ;-) jk

GaryDoug Posted at 11-19 20:40
Wow that was going as you said "like a bat out of hell" with no forward right stick. And how does a Mini 2 go 45 mph?.... with a big wind maybe. Lab will probably come up with the best analysis...or guess ;-) jk

Yep. It was zipping along for sure.

DGCA3 Posted at 11-19 20:43
Yep. 45 miles per hour. It was zipping along.

Well, my point was that it is beyond it's capability...at least in the specs. Maybe some kind of instrumentation fault?

On another issue, maybe 817 feet is a bit high for an RTH altitude ;-)

GaryDoug Posted at 11-19 20:45
Well, my point was that it is beyond it's capability...at least in the specs. Maybe some kind of instrumentation fault?

Your guess is as  good as mine.

GaryDoug Posted at 11-19 20:45
Well, my point was that it is beyond it's capability...at least in the specs. Maybe some kind of instrumentation fault?

On another issue, maybe 817 feet is a bit high for an RTH altitude ;-)

"On another issue, maybe 817 feet is a bit high for an RTH altitude ;-)"

Yes indeed. Not sure why I set it at that. Must of confused it with distance or max height on one of my previous flights and just left it as is.

Hello there DGCA3. I am sorry to know what happened to your previous flight. Can you please fly your DJI Mini 2 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 on the ground, and you took off from close to a buildings with rebar concrete metal objects on it. Thank you.

DJI Stephen Posted at 11-19 20:56
Hello there DGCA3. I am sorry to know what happened to your previous flight. Can you please fly your DJI Mini 2 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 on the ground, and you took off from close to a buildings with rebar concrete metal objects on it. Thank you.

After this happened, I flew the drone back over to above me, and then flew it out over the gravel plant without any other issues. I've also flown it closer to electrical lines in previous flights (not on purpose)  with only the brief interference notification but no effect on my connection or flight.  It's the total randomness of this incident that concerns me. If you look at the end of the video, there's really nothing around me. It should be the ideal location to launch and land from.

DGCA3 Posted at 11-19 20:30
It's the first time I've accessed this stuff. I'm a newbie with data.

Here's what that flight data looks like:
https://www.phantomhelp.com/logviewer/15VUXY3GWHPKWOO7HDCZ/#

At first glance it doesn't fit the common yaw error pattern of a smoothly curved flight path but it does have the rapid uncommanded acceleration beyond the spec top speed.
The flight data does show a subtle uncommanded heading change starting around 15.1 sec which coincides with the drone starting to slowly accelerate from zero.

Matching the data to the drone's movement is complicated by the use of Sport Mode and full stick inputs in opposite directions.
This makes it tricky to tell what movement was the mystery incident and what was joystick driven.
It appears that the drone regained stability around 24 sec or maybe earlier.

I don't like to guess about what there's no evidence for so I'd still like a description of the launch surface and what was within 10 feet of it.
Also confirmation that you placed the drone at the launch point and powered it on where it was launched from.
And what direction was the drone facing at the launch?

This incident is puzzling and might suit the analysis technique of JJB*

Labroides Posted at 11-19 21:40
Here's what that flight data looks like:
https://www.phantomhelp.com/logviewer/15VUXY3GWHPKWOO7HDCZ/#

Thanks for the detailed reply.

The street surface was asphalt. Just me and my car were nearby. If you watch the end of the video, you'll see there is nothing else around. I believe the drone was facing mostly south when taking off. Also, the log doesn't show that it got lower than that at the end of the errant run. Looks like 5 feet or so, on it's way to hitting a wall before recovering altitude.

Oh the dreaded fly away still exists in the mini 2, fun.

Remember this video with the mini 1? I've seen it happen with nothing around, I've seen it Yeet full speed after hitting something, Be thankful you were able to regain control. I know when my mini took off for the moon I didn't regain control until it was to late, still odd the mini behaves in such a manor.  

DGCA3 Posted at 11-19 22:04
Thanks for the detailed reply.

The street surface was asphalt. Just me and my car were nearby. If you watch the end of the video, you'll see there is nothing else around. I believe the drone was facing mostly south when taking off. Also, the log doesn't show that it got lower than that at the end of the errant run. Looks like 5 feet or so, on it's way to hitting a wall before recovering altitude.

If it really was pointing south, that strongly suggests a yaw error incident because the data shows it pointing just west of north.

Labroides Posted at 11-19 22:12
If it really was pointing south, that strongly suggests a yaw error incident because the data shows it pointing just west of north.

I just looked at the video from when it actually left the asphalt. It was right after I did a compass calibration. It was pointing at the gravel plant, which was mostly south, maybe just a little east.

Post your log at http://phantomhelp.com/logviewer/ it will show at lot more information.

DAFlys Posted at 11-20 01:47
Post your log at http://phantomhelp.com/logviewer/ it will show at lot more information.

That's already taken care of .. see my large post above.

Labroides Posted at 11-20 02:01
That's already taken care of .. see my large post above.

Cool - I missed that.   thx

Magnetic interference by the look of it

It sounds a bit suspicious, Look at post from "fansd58737bb" Mini 2 crashed 2nd flight, it looks near identical to this near crash!
We need someone like JJB to compare the flight logs of these 2 incidents, to see if there are more reasons why this happened.

videoeditman Posted at 11-20 06:03
It sounds a bit suspicious, Look at post from "fansd58737bb" Mini 2 crashed 2nd flight, it looks near identical to this near crash!
We need someone like JJB to compare the flight logs of these 2 incidents, to see if there are more reasons why this happened.

By saying it's 'suspicious', what are you implying?

Both drones hit speeds that the mini can't attain, both Yawed wildly with no input. I would say that whatever caused one issue also caused the other. We need to get to the bottom of this before more doomed flights happen. Just saying it doesn't seem to be totally random as there seems to be a very similar problem between these 2 incidents.

videoeditman Posted at 11-20 06:31
Both drones hit speeds that the mini can't attain, both Yawed wildly with no input. I would say that whatever caused one issue also caused the other. We need to get to the bottom of this before more doomed flights happen. Just saying it doesn't seem to be totally random as there seems to be a very similar problem between these 2 incidents.

Okay, I understand what you're saying.

I'm glad your drone came back to fly another day, but I bet you are a bit nervous now when flying!

Labroides Posted at 11-19 22:12
If it really was pointing south, that strongly suggests a yaw error incident because the data shows it pointing just west of north.

You can see in the video that it starts off pointing south-southeast (towards the plant), but the data shows the opposite, pointing north-northwest. Do we think the op took off from the roof of the car? I see he is standing right beside the car. Or metal under the cul-de-sac pavement?

Hi,

Had a look at your flightlog too.

Without your DAT file for this flight i am only 98% sure....

See the polar chart ; yellow line is move line without RC Pitch or Roll input (so it should steady hover)
In the other chart see the YAW line and the YAW reference line. in normal flights they are the same..not always in heading values but the offset to another stays the same....

Not at your start....further in flight yes.

This is voor 98% a compass induced fly away, happily a short one.

If you like upload your  DAT file for this flight (also on your mobile device in the MCDatFlightRecords map)

ALWAYS check your compass arrow in the map view ; check if arrow is pointing same heading as  your craft in the real world....do this before taek-off.if not the same ; DO NOT start a flight.

PS craft heading at start 340 in the log, not sure but checking the take-off position (road direction) on the map ;   actual heading not 340.
cheers
JJB