A few days ago I had a bad experience as the mavic mini as it goes erratic without my control input.- Did all the pre-flight checks all good.
- No compass or IMU calibration required.
- Fly permitted shown in the monitor.
- >9 Satellite & signal all good.

After I flew normally for 2 minutes, the next 5 mins the mavic mini goes totally crazy without my control input, fortunately the drone managed to return to home base by itself when it loses signal. Lucklily during these moment when i had no control, it does not crash into any objects.

Could someone from DJI techncial personnel look into my flight log and decipher what is happening?

Hello there. I am sorry for the trouble this has caused and thank you for reaching out. Kindly please try calibrate the remote controller of your DJI Mavic Mini to see if the issue will persist. please keep us posted for further assistance. Thank you.

DJI Stephen Posted at 5-22 21:20
Hello there. I am sorry for the trouble this has caused and thank you for reaching out. Kindly please try calibrate the remote controller of your DJI Mavic Mini to see if the issue will persist. please keep us posted for further assistance. Thank you.

Thank you for your feedback.

What if the calibration successful and eventually does not work during the flight test, what if it goes erratic again. This time I might not be so lucky to get the Mavic Mini in one piece.

Do you want to have a look at my flight log to at least determine the root cause?

djiuser_FNeBvQNjxql3 Posted at 5-22 21:56
Thank you for your feedback.

What if the calibration successful and eventually does not work during the flight test, what if it goes erratic again. This time I might not be so lucky to get the Mavic Mini in one piece.

It's unlikely that calibrating things would make any difference.
It would make sense to identify the problem and address that before calibrating anything.

Go to: https://www.phantomhelp.com/LogViewer/Upload/
... where you'll find instructions.
That will give a report you can post a link for the report here.
Or just post the .txt file.

When someone suggests recalibrating things without seeing the flight data, that's a good indication that they have no idea what the problem was or how to fix it.
The first thing to do in solving flight problems is to identify the problem and to do that, the recorded flight data is usually needed.

Here's what that flight data looks like:
https://www.phantomhelp.com/logviewer/5H67HAC0FSY4F0XQ1PLI/

About 4 minutes into the flight (around 8:40) the speed becomes erratic and the drone starts flying a smooth oval course without matching joystick inputs.
This has some of the indications of a yaw error incident.
You swapped flight modes to Sport and Tripod but the drone continued flying without control.

You initiated RTH at 9:34.9 and it was able to bring the drone home.
It looks like you were lucky and the drone had a small angle yaw error.
With a larger angle, the yaw error is usually serious and the drone is lost or crashes.
With a small error, the drone doesn't go as fast and RTH can sometimes work.
But with a smaller angle, the yaw error is more difficult to properly identify.

With a yaw error incident there is nothing wrong with the drone, but the problem is launching from a magnetically "dirty" area.
It looks like the launch point was an a building roof, perhaps steel or reinforced concrete?
Do you remember which way the drone was pointing while waiting for launch?

Labroides Posted at 5-23 04:04
When someone suggests recalibrating things without seeing the flight data, that's a good indication that they have no idea what the problem was or how to fix it.

Here's what that flight data looks like:

Hi Labroides, thank you for the clear explanation.

1// The launch was at the roof with no obstructions or frame work above. I believe the building itself was made with reinforce concrete.

2// The initiation launch the drone was facing south as it shown in the flight plan.

3// Is there a possibility that someone was using a radar jammer? or someone used the same frequency as i do?

4// Is there any ways to avoid such a situation?

djiuser_FNeBvQNjxql3 Posted at 5-23 04:12
Hi Labroides, thank you for the clear explanation.

1// The launch was at the roof with no obstructions or frame work above. I believe the building itself was made with reinforce concrete.

1// I believe the building itself was made with reinforce concrete.
It's never a good idea to launch from reinforced concrete surfaces.
Your compass was deflected and the gyro sensor in the IMU picks up this incorrect heading value when it initialises.
When you launch and get far enough from the concrete, the compass goes back to reading correctly, but the gyro sensor data now disagrees with the compass and the problems start.
The drone senses that it has moved off position and tries to correct.
But using the incorrect directional data from the gyro, it gets further off course, speeds up and corrects further, getting further off etc, etc.

When the difference between compass and gyro headings is small, the corrections are less severe and the flight movements are less drastic.
With larger differences, the drone reaches very high speeds and RTH would have no chance to bring it back.

2// The initiation launch the drone was facing south as it shown in the flight plan.The difference in angle between the actual heading and the false heading was probably too small to easily see.
With larger angles, it is obvious with the drone facing something like 90 degrees away from the direction shown in the data

3// Is there a possibility that someone was using a radar jammer? or someone used the same frequency as i do?
No and no

4// Is there any ways to avoid such a situation?
Don't launch from reinforced concrete surfaces or within a metre or so of steel objects.
If you want to launch from somewhere like that, try using a plastic or wood box etc to give separation between the drone and the steel.

If you get a compass error from a launch surface, the compass is warning of a magnetic interference problem, because the magnetic influence is stronger than the arth's normal magnetic field.
Don't recalibrate the compass, but switch off and move away and launch from somewhere else.
If the magnetic field is of similar strength to the earth's, you might get no warning, but because the field is a different direction from the earth's, then you can get a yaw error.
Most are large angle errors and easily detected by checking that the drone is pointing the same compass direction as the drone icon on the map window.