What happened?
- Thread starter Javierz0509
- Start date
Yep ... you had a uncommanded horizontal movement which reached 7,3mph (3,3m/s) at max up on 292ft (89m) height. HP was properly recorded & both GPS lock & GPSLevel was good.
The movement is clearly seen here ... have placed the chart marker at 30,1sec where the uncommanded movement is fastest. No horizontal stick inputs where made (aileron=roll=value 1024 neutral & elevator=pitch=value 1024=neutral)
(Click on the pictures below to make them larger)
But going in and comparing the GPS velocity with the IMU velocity in direction North & East shows a large deviation there (should be very near to 0 mph) ... this usually indicate a yaw error coming from powering on the craft in a magnetic interfered location ... or a IMU error.
This velocity difference is also seen in below sat.picture where the red path is from the GPS & the green is what the IMU thought. You can also see that the craft was recorded to be pointing in a south easterly direction just at take off ... was that the case ... or did it point into another direction in reality?
The movement is clearly seen here ... have placed the chart marker at 30,1sec where the uncommanded movement is fastest. No horizontal stick inputs where made (aileron=roll=value 1024 neutral & elevator=pitch=value 1024=neutral)
(Click on the pictures below to make them larger)
But going in and comparing the GPS velocity with the IMU velocity in direction North & East shows a large deviation there (should be very near to 0 mph) ... this usually indicate a yaw error coming from powering on the craft in a magnetic interfered location ... or a IMU error.
This velocity difference is also seen in below sat.picture where the red path is from the GPS & the green is what the IMU thought. You can also see that the craft was recorded to be pointing in a south easterly direction just at take off ... was that the case ... or did it point into another direction in reality?
thank you for taking the time to look into this for me, it was a scary moment when no input is given and the drone does stuff on its own, here is a video of what happened I will be calling DJI tomorrow to see if they can replace the drone since still has warranty and DJI refresh .
just uploaded the video so the quality might be bad.
Yaros
Well-Known Member
Try to recalibrate the IMU and Compass.
Then try again in an open location
Then try again in an open location
This is often recommended by people in forums, but usually when neither would make any difference.
This is another of those occasions.
Recalibrating the compass and IMU are not needed and won't make any difference.
Yaros
Well-Known Member
Yeah ... your clip pretty much confirms a yaw error.
This was the view & the direction your craft was pointing in reality at take off ...
In below pic. the black arrow shows the real heading direction ... the red the direction the IMU thought.
The yaw error is way over 90 degrees ... & that would create a kind of circular "drift" path as shown in the log.
Without access to any decrypted .DAT log it will be impossible for us to reveal the root cause here unfortunately.
Calibrating the compass & IMU will not fix a yaw error ...
There's no need to do it before the first flight.
That would be mentioned in the manual if it was important.
And it's always better to identify the cause of the issue and address that rather than just doing things because "it doesn't hurt to try".
Would you think it was a good idea if your mechanic or doctor tried that approach?
- Joined
- May 25, 2017
- Messages
- 8,476
- Reactions
- 7,963
- Age
- 62
Checking the actual drone orientation on the ground to what is shown on the map is one of the most important pre flight checks you can do.
If that is out, you move take off point somewhere nearby, where metal type compass interference is not present (or hand launch might fix it).
At least it didn't go full steam ahead 90 degrees and crash into a building . . . this time.
If that is out, you move take off point somewhere nearby, where metal type compass interference is not present (or hand launch might fix it).
At least it didn't go full steam ahead 90 degrees and crash into a building . . . this time.
No ... obstacle avoidance doesn't cause yaw errors.
In this case the AC was pointing in a Northerly direction in reality confirmed by the clip the OP shared. The IMU though ... thought the AC was pointing in a south easterly direction, confirmed by the log... you have it all in post #7.
All this is fine as long the flight controller doesn't need to counteract anything (like the wind) in order to hold position. But as soon as it's a need to correct the position the flight controller will increase the thrust with the wrong motors (due to not having the correct heading direction) ... this will instead of getting back to the correct position, increase the positional error ... & the loop continues with more & more thrust applied with the wrong motors until the AC flyaway in full uncontrollable speed.
This time the OP were lucky & could descend down before the full consequences of the yaw error set off ...
Just from the pictures, there seem to be a lot of metal structures around. Also, it looks like you launched from near the edge of a parking lot. Might you have been over a buried metal drain pipe?
Based on the flight records that you posted there is nothing wrong with the aircraft - you simply powered it up in the presence of an external interfering magnetic field and didn't do the basic check to see if the heading indication agreed with the direction that it was pointing on the ground before takeoff.
If you want a more detailed description of that problem:
A short explanation of compass function, calibration and errors.
BACKGROUND The aircraft compass is often misunderstood, both in terms of function and purpose, but it is important to understand what the compass measures, why, and what happens to that information. At the most basic level the purpose of the compass is to measure the earth’s magnetic field in...
mavicpilots.com
Based on the flight records that you posted there is nothing wrong with the aircraft - you simply powered it up in the presence of an external interfering magnetic field and didn't do the basic check to see if the heading indication agreed with the direction that it was pointing on the ground before takeoff.
If you want a more detailed description of that problem:
A short explanation of compass function, calibration and errors.
BACKGROUND The aircraft compass is often misunderstood, both in terms of function and purpose, but it is important to understand what the compass measures, why, and what happens to that information. At the most basic level the purpose of the compass is to measure the earth’s magnetic field in...
the reason I am more concerned is cause where i live as i was going forward and pulling up at the same time the drone started drifting to the left no obstacle around it.
As said, a yaw error isn't anything permanent if it's due to a power on in a magnetic interfered location ... & isn't then a sign of a faulty craft.
When you power up your craft the IMU heading direction is initialized by the compass ... if this happens in a magnetic interfered location the compass will be deflected & by that the IMU will get the wrong information regarding how the craft is pointing.
As soon as the craft ascends away from the magnetic interfered location the compass will correct it self but not the IMU, which mainly is what tells the flight controller in what direction the craft points during the flight.
All this with a wrongly initialized IMU is easy to see before getting airborne ... just look at the map view in the FLY app & the drone icon there ... if that icon is pointing equal there in relation to other map object as the drone does in reality ... all is good. If not ... you need to power the craft down, move away to another take-off location & power up & repeat.
Last edited:
before that flight I did take off from a park away from that area and the drone perform really well.
here is the flight information before I took off from the downtown area.
Might add in this ...
As the .DAT logs from a M3 can't be decrypted & read by us we can't say if this yaw error was due to a magnetic interfered power on location (can be handled by the pilot) ... or a IMU error.
If you follow the pre-take off check on the map as I proposed in my previous post & everything looks good but the craft behaves anyway ... it CAN be due to a faulty IMU.
If your craft was pointing in this direction at take-off ... you didn't have a yaw error then.
Similar threads
