Take off from boat - word of warning

[Meta4]

View attachment 110490
It did eventually go into RTH mode where the yellow arrow is located and returned to the boat where i eventually reclaimed manual control of it. how do i get the DAT file?

The yellow line is an imaginary line joining the scrambled parts of the data from the original file.
The drone never flew that straight line.

To see where the "RTH" section of the flight was, click on the link in your post #47 and jump down to around 9:16 where it shows Go Home as the flight mode, and click on the line.
The red drone icon shows where the drone was at that time.
The drone was trying to RTH but as the curved path shows, it wasn't able to.

 

[joeyfri13]

The yellow line is an imaginary line joining the scrambled parts of the data from the original file.
The drone never flew that straight line.

it might not have been exactly straight but it did return to the boat from that location in RTH mode you can see it in the video. 10:21 it looks like it finally goes into RTH mode.

 

[boblui]

The airdata and the footage played side by side. In the later part of the video, the craft thought that it was flying backward but the footage showed the opposite :

 

[BudWalker]

View attachment 110490
It did eventually go into RTH mode where the yellow arrow is located and returned to the boat where i eventually reclaimed manual control of it. how do i get the DAT file?

To get the .DAT go to

Mavic Flight Log Retrieval and Analysis Guide

With the introduction of the Mavic Mini there has been a noticeable uptick in the number of posts requesting help with lost or crashed aircraft. While there is plenty of help and advice available on this forum, and detailed guidance and options on several other websites, I thought that perhaps a...

mavicpilots.com

Section 3, subheading Aircraft DAT files

 

[joeyfri13]

To get the .DAT go to

Mavic Flight Log Retrieval and Analysis Guide

With the introduction of the Mavic Mini there has been a noticeable uptick in the number of posts requesting help with lost or crashed aircraft. While there is plenty of help and advice available on this forum, and detailed guidance and options on several other websites, I thought that perhaps a...

mavicpilots.com

Section 3, subheading Aircraft DAT files

Thank you, I will get on it when I get home from work.

 

[joeyfri13]

I think I am beginning to understand the more I read. So initially everything was fine I had it hovering behind the boat for approximately 9 minutes, I made a last adjustment to line it up with the back of the boat because of a combination of boat sway and Wind drift. For unknown reasons there is some magnetic interference and then the mini does not know which way it is facing causing it to fly towards the bank I was able to turn the drone back over the water only to lose control again and have it dip hard left towards the water, I lost my video feed at that point but I held down the RTH in a panic. I thought the drone was in the river as I lost sight of it and could not heard it because of a passing boat. My buddy looks up to find it descending rapidly. I manage to take control to land it in my hand at that point. I am curious and slightly worried about what would have caused the magnetic interference because making videos while fishing is one of the main reasons I purchased it in the first place. Hopefully when I get the dat file it will shed some light.

 

[boblui]

.... I am curious and slightly worried about what would have caused the magnetic interference

It is highly unlikely that there was magnetic interference over the middle of a river. The problem was caused either by a glitch in the gyro sensor or, as seen in a recent case, the flight controller made a big wrong adjustment to the yaw value for unknown reason.

The fact that you were able to regain control is due to the yaw get corrected automatically after the wild run. At this point, the footage shows that the drone was pointing to the bridge, so is the indication of the yaw :

 

[joeyfri13]

The yellow line is an imaginary line joining the scrambled parts of the data from the original file.
The drone never flew that straight line.

To see where the "RTH" section of the flight was, click on the link in your post #47 and jump down to around 9:16 where it shows Go Home as the flight mode, and click on the line.
The red drone icon shows where the drone was at that time.
The drone was trying to RTH but as the curved path shows, it wasn't able to.

yes i understand that, however if you watch the entire video you can see where the drone eventually corrects it self adjusts to my preset altitude and returns to the boat which is a straight line from the end of the curve over the river. Somehow the yaw glitch self corrected, the data is mixed up

 

[Meta4]

yes i understand that, however if you watch the entire video you can see where the drone eventually corrects it self adjusts to my preset altitude and returns to the boat which is a straight line from the end of the curve over the river. Somehow the yaw glitch self corrected, the data is mixed up

The actual path is shown in post #56.

 

[BudWalker]

yes i understand that, however if you watch the entire video you can see where the drone eventually corrects it self adjusts to my preset altitude and returns to the boat which is a straight line from the end of the curve over the river. Somehow the yaw glitch self corrected, the data is mixed up

Yes, the data is mixed up. The .csv that was generated by TXTlogToCSVtool was out of order. As I said in my post #56 I manually re-ordered the .csv. Unfortunately, it was not only out of order it had the last part of the flight missing. That's why it doesn't show the RTH portion of the flight. But, this missing part isn't relevant to the cause of the fly away.

The captured part of the flight does show that RTH was in effect from 553 to 583 secs. I'm guessing that you probably re-pressed the Go Home button because the MM wasn't returning. That's why RTH terminated at 583 secs. The second and successful RTH wasn't captured.

Have you been able to retrieve FLY054?

And, if you haven't powered up the MM can you retrieve a file from the external SD card that has the video you submitted. The file is MISC/LOG/flylog/fc_log.log. It get's overwritten every time the MM is powered up.

 

[joeyfri13]

Yes, the data is mixed up. The .csv that was generated by TXTlogToCSVtool was out of order. As I said in my post #56 I manually re-ordered the .csv. Unfortunately, it was not only out of order it had the last part of the flight missing. That's why it doesn't show the RTH portion of the flight. But, this missing part isn't relevant to the cause of the fly away.

The captured part of the flight does show that RTH was in effect from 553 to 583 secs. I'm guessing that you probably re-pressed the Go Home button because the MM wasn't returning. That's why RTH terminated at 583 secs. The second and successful RTH wasn't captured.

Have you been able to retrieve FLY054?

And, if you haven't powered up the MM can you retrieve a file from the external SD card that has the video you submitted. The file is MISC/LOG/flylog/fc_log.log. It get's overwritten every time the MM is powered up.

I started to work on that but got sidetracked with family stuff so I will finish getting the dat file uploaded today. The drone was already powered up since then because I was unaware of the availability of this file at the time. Now I know if they happens again not to power it up again until I have extracted that file from the memory card.

 

[joeyfri13]

I attempted to get the dat file but there is nothing in the sub folder, I’m guessing DJI deleted it when they uploaded to the server?

 

[calltrex]

I take off our pontoon all the time. Never had any hiccups yet. I always put the auto floater on it incase its goes in the water

 

[Flytelandon]

Here is the video of the fly away. Launch at about 00:16

How long did it take you too find the mm, since you were on the water? Look like it travelled a distance

 

[hydrolift]

How long did it take you too find the mm, since you were on the water? Look like it travelled a distance

Hi!
It didn't travel too far. After getting off the boat I found it after only 5 minutes. But it was in the middle of some thick thorn bushes so it took a while to actually reach it.

I change the propellers today and tested it on a soccer field. It worked like a charm which was a relief!

But I remember something that I reacted to a few days ago. A small glitch that can be found in this video (don't know if it is related to the incident on the boat):

PS. Must be the gimbal and not the drone FC, right?

 

[moonlitnite]

I'm of the mind that all the iron in the outboard motor caused the compass initialization to fail (I didn't notice you saying you did the "compass dance" before takeoff). Going back to my Phantom 1 and 2s, I quickly learned the painful lessons from others to NEVER initialize and take off near a metal surface of size. For instance taking off on the top level of a parking garage with rebar in the concrete is a big no-no. Ditto taking off on a ferromagnetic table, etc. You might get by taking off from the roof of a car but I wouldn't risk it. Others have also learned that taking off from a cruise ship requires hand launching or at least taking off on a non ferrous table several feet above the steel deck.

Thanks for sharing!

Happy trails, Michael

 

[Meta4]

I'm of the mind that all the iron in the outboard motor caused the compass initialization to fail (I didn't notice you saying you did the "compass dance" before takeoff).

If there are magnetic problems at the launch point, no amount of compass dancing will solve them.

 

[moonlitnite]

Meta4, you're certainly entitled to your opinion. IMO when holding the UAF 3 feet above a ferrous material, the compass will properly calibrate. It's not as though getting near a metallic object after calibration will reinitialize the compass and IMU.

I think you're confusing a ferromagnetic material with induction . In the case iron or steel, it doesn't emit a magnetic field of sufficient Gauss such as 3,000-7,000G from a MRI machine which certainly would likely decalibrate a UAV within perhaps 15 feet.

Regardless, I'm sure we'll all agree to error on the side of caution, if not a bit of superstition (and yes, I still do the "compass dance" more often then necessary).

 

[boblui]

... A small glitch that can be found in this video (don't know if it is related to the incident on the boat):

PS. Must be the gimbal and not the drone FC, right?

My Mini has done that a few times in the same flight. It was flying forward at full speed in sport mode. With no rudder input, the craft suddenly yawed like what's shown in your video. Upon checking the flight logs, it was not the gimbal but the craft has yawed.

 

[hydrolift]

My Mini has done that a few times in the same flight. It was flying forward at full speed in sport mode. With no rudder input, the craft suddenly yawed like what's shown in your video. Upon checking the flight logs, it was not the gimbal but the craft has yawed.

View attachment 110713

Wow! Another scary thing with the MM. Not as safe to fly as I thought when I bought it...

Did you have any incidents following the glitch? I was thinking the 'yawing behaviour' might have led to something more serious?