Mavic 3 flies forward (on its own) crashing into a house wall

[BudWalker]

Not just a derivative of position data, rather than integral?

I'll have to look up Kalman filters and do some reading.

Thanks.

Yes, of course, you are correct. I'll fix that error.

 

[Meta4]

So you do not have a theory ...

I explained that in post #47

and everything else in your post is simply wrong/ incorrect.

That is blatantly wrong.
I put effort into solving your complicated issue, despite having no idea what the flying environment was because you deliberately hid information.
I won't bother again.

 

[Mordin]

@Meta4
I appreciate the things you explained/ mentioned about the log, I really do.
But when you say "Your drone did respond to your control inputs, just not enough to save teh drone from crashing into the obstacle tou pushed it towards." ... That sounds like "I directed the drone to fly towards a wall ... then I tried to reverse course, and it was too late ... so it crashed - PIC's fault".

Do you mean the nudge I gave the drone when it was 10ft away from the wall? That was my "pushing it towards" the wall?
You do know that I let go of the control sticks when the drone was 8-10ft from the wall, right? At that time the drone was slow (1ft/ sec. or so). Then it accelerated on its own toward the wall ... and during that flight-path all inputs from me (via control sticks) were to move the drone backwards/ away from the wall.

Please explain how that is my fault? ... or how it is an issue of 1 or 2ft.?
The drone accelerated on its own when it was 8-10ft from the wall. I am not sure why you try to "forget" that fact ... or do you not believe it?

Or do yo mean it like this: the drone was flying toward the wall (let's not go into details why), then the controls were applied to go backwards - the drone slows down, and would have stopped if the wall had been 1 or 2ft further away.
That would be an "interesting" approach of saying it is an issue of 1 or 2ft.

 

[Mordin]

if the .csv is modified by offfsetting the latitude and longitude values then it's possible to obtain the original values.

How would that be possible???
If I add 1.234 degrees to every latitude (or longitude) entry in my log (that would shift my location by about 120km) ... how can someone figure out my actual location? They don't know if I added 1.5 or 0.789 ...

And don't get me wrong - I am not some math-genius ... but for a lot of things I understand the theory, and someone with thorough math abilities could calculate it.
For this I have no idea how someone would start to find my actual location,

For the longitude I understand the actual distance gets shorter for a specific angle the higher up you are on the globe. Maybe with the Speed values you can get an idea of where the person originally was ... but even there with all the inaccuracies of Speed entries ... I don't see it.

I am very interested what the theory behind these calculations is.

 

[Meta4]

Please explain how that is my fault?

I've wasted more time on this thankless task than I care to already

 

[pentatony]

Why?
There's nothing to indicate you had any GPS issue.

Learn not to fly so close and toward obstacles.
Find a nice clear area, well away from obstacles to do your flying.

You would have to have a case where you could show that there was an actual issue DJI was responsible for.

I'm still learning not to fly to close to things. Over time, the little bumps, drops or minor crashes will cause some sort of drone malfunction that can only be repaired by professionals.

 

[Mordin]

I'm still learning not to fly to close to things. Over time, the little bumps, drops or minor crashes will cause some sort of drone malfunction that can only be repaired by professionals.

I am also still learning how to fly properly ... but about a year ago I started to fly lots of "orbit" and other maneuvers around trees and other objects ... where the orientation of the drone constantly changes. It did help me a lot to avoid crashes when something unexpected happens (wind gust, or the flight path of the drone does not go exactly as planned ...) and the sticks have to be moved in a correcting way to avoid a crash.

That's why it is so frustrating to me that my latest/ newest drone decides to fly on its own toward a wall, and even with correcting actions on the sticks it was not possible to avert the crash.

 

[Mordin]

I've wasted more time on this thankless task than I care to already

Ok,
Thanks for your help!

 

[lolo780]

I lost a Mini 2 when the navigation system decide to flip direction 180deg. Pulling back on the stick drove it into a cliff at full speed.

 

[Meta4]

I lost a Mini 2 when the navigation system decide to flip direction 180deg. Pulling back on the stick drove it into a cliff at full speed.

That sounds dubious.
Did you have any analysis done on the flight data?

 

[lolo780]

That sounds dubious.
Did you have any analysis done on the flight data?

The heading on the map pointed away from the cliff after takeoff even though the camera recording showed heading was towards the cliff.

 

[Meta4]

The heading on the map pointed away from the cliff after takeoff even though the camera recording showed heading was towards the cliff.

Your incident description sounds unlikely.
Did you have the flight data analysed or is it just something that you guessed.

 

[slup]

The heading on the map pointed away from the cliff after takeoff even though the camera recording showed heading was towards the cliff.

If this state was apparent already at take off, it most probably was a ordinary "yaw error" caused by you powering on the drone in a magnetic disturbed area.

Something deflected the compass 180 degrees when the drone powered up & that initialized the IMU in the wrong direction. All this is clearly detectable before taking off by just looking at the map view in the app & the drone icon there... if it's not pointing equal in relation to other map objects as it does in reality relative the same objects, you need to power down & move away to another spot, power up & check again.

If this state suddenly happened mid flight it could point towards a rare IMU error...

 

[lolo780]

Mini 2's are disposable junk and I never looked at more than the app replay. I had to pick this one up with a Mavic 3 once after landing on a metal roof and the stupid Mini 2 refusing to start the motors due to compass error.

 

[Meta4]

Mini 2's are disposable junk and I never looked at more than the app replay. I had to pick this one up with a Mavic 3 once after landing on a metal roof and the stupid Mini 2 refusing to start the motors due to compass error.

The drone doesn't "decide to flip direction 180deg"
If you had asked for analysis of the data, you could have found out what caused the incident and how to prevent a similar occurrence in future.