Like the title says, I experienced something extremely weird, and potentially very dangerous on 4th of July. Just to be clear, my drone is fine, this is not one of those "my drone broke and its YOUR fault!" kind of thread. But it very well could have been.

First som backstory. I crashed my drone about a month ago. It was my own fault, and I paid for repairs at the California DJI Repair Center. I received my drone back (or a swap unit) on June 27th. I paired it with my existing controller successfully, and had a few uneventful flights after this.

Fast forward to 4th of July. I am about to film the fireworks in Jacksonville, FL. I am alone at the time, but I have a friend who also own a Mavic Pro, who did the same thing. We were about a quarter of a mile from each other, and didnt realize what (possibly) have happened until we both shared our stories a few days after the incident(s).

MY STORY:
I take off with 9 satellites locked, and a "ready to go" message on my controller. About 1 minute into my flight, I get a compass error, immediately followed by a IMU Heading exception (in flight) Please switch to heading mode if aircraft behave abnormally). At this time I am about 70 meters from my position, and at an altitude of 45 meters. Fortunately I look up to regain a visual position on my drone, and I see it moving at full sport-mode speed to the side, without me doing any input at all. I input the opposite inputs that what it is doing to try and counter its movement, as it is flying towards a building. Fortunately, it stops, and presumaly hovers just fine. I fly back to my own position, land, restart everything, and take off again. For this flight, everything behaved normally.

At first, I assumed this was because I "only" had 9 satellites connected, and somehow it was not enough to get a good solution, but after hearing my friends story, I now believe differently:

HIS STORY:
He took off about the same time as me (I have not seen his flight record of the flight yet, so some of this is assumptions for now). While flying, he was looking at the video feed, and suddendly it started to become distorted, as if the signal was interfered, or weak. His distance to his drone was not much more than at most a few hundred meters at this time. Suddendly, he sees a COMPLETELY different picture on his display, as if he is watching a different video feed. It is similar of course, but from a very different position (which coincides with the fact that I was in a different position). This is confusing for him, so he tried to maneuver the drone, but since the video feed does not match up with the position of his drone, he loses his visual position of his own drone in the confusion. At this time, he looks back down at the video feed and tried to locate his positon by video or by the map. He is unable to do this immediately, and makes inputs to the controller to move the drone to get a better view, or recognize some feature to place his position.

After making these inputs, he suddendly loses the video feed, and it jumps back to his own drone feed. The story continues after this, but I am going to skip those parts for now, as they are not relevant to the topic of this thread:
tl;dr
Another RC disconnected from his own drone, and connected to mine, in air, without being previously paired to my drone.


Have any of you ever experienced something similar to this? And I assume this is not supposed to happen? Does DJI have any comment, or would like to review our records to see what has happened?

Interesting. I'd like to take a look at the logs from the flight. Can you please upload the TXT file from your mobile device here and the DAT file from the Mavic to DropBox using these instructions and provide a link to both in this thread?

Paranormal Activity !.

That's a real YIKES!
Glad both of the drones survived.

There has to be some other explanation because I just cannot see how the drone could pair with another RC. Something strange might have occurred if either one of you inadvertently tapped on secondary controller but even that is a long stretch.

You should both pair your controllers again.

not sure how much of that my heart could take - hope u get it resolved soon

I don't think that's possible, but there may have been interference that caused this behavior...

As others have stated...I don't think that its possible without physically pairing the RC to the drone.

I'll upload the logs when I get the chance. And no, I don't really think it is possible either. I have seen my logs like I said, but not my friends logs. So I'm not positive he actually saw another feed on his screen, but that is what it appeared to be. It could just be a loss of situational awareness on his side, and some IMU/GPS error on my drone. I do know that my drone flew at full speed in one direction, without any inputs on my controller though.

.txt file:
http://www.phantomhelp.com/LogViewer/T26QQVFCPE0GPLZ3AANV/

I may have jumped the gun a little, after hearing my friends story, and how the two timeframes lined up perfectly, but it is possible that it was "just" an internal error in my drone. I never saw the "yaw error" on my display when it happened, or in playback, but there is a lot of them in the text file.

My dropbox is full, so I had to upload the .DAT file to google drive. Hope that is fine.

https://drive.google.com/open?id=0B2owTs8nOJs9Snd0SkZZNGRuaDg

Interesting. I'm not sure if there is enough evidence to reach your conclusion but everything seems to be pointing that way!

I've seen stranger things happen, any things possible...

phantom_of_the_ Posted at 2017-7-9 10:26
Interesting. I'm not sure if there is enough evidence to reach your conclusion but everything seems to be pointing that way!

Including the logs? I dont know how to look at the DAT file...

Christopher90 Posted at 2017-7-9 07:36
My dropbox is full, so I had to upload the .DAT file to google drive. Hope that is fine.

https://drive.google.com/open?id=0B2owTs8nOJs9Snd0SkZZNGRuaDg

Yep, that works. I looked at the DAT file and while I can't definitively say why you got the errors you did, I can say with a high degree of confidence that this wasn't caused by interference between your system and that of your friend.

Essentially what happened is that you wound up in ATTI mode due to unexpected compass readings during flight. That most commonly is caused by taking off from a geomagnetically distorted location but that doesn't appear to be what happened here. In fact, your compass errors didn't show up until about 20 seconds into the flight when you were well away from the ground and the building near which your takeoff occurred.

Basically what the graph below shows is that your compass heading -- "magYaw" represented by the blue plot -- starts gradually increasing at around 0:20 but your expected heading (yaw represented by the purple line) stays essentially unchanged. The magYaw is sometimes the more reliable heading, but based on the direction of flight shown in the log it appears that yaw was the more accurate value here. The yaw value also makes more sense given that you weren't using the stick to turn the aircraft at that point (the "rudder" value indicated by the beige line), nor is there any significant rotation detected by the gyro (the red line).

Once yaw and magYaw diverged by a significant amount at around 0:21, the flight controller (FC) triggered ATTI mode indicated by the purple region on the graph, and the rest of what you experienced is pretty normal behavior in these circumstances.

So really the only question here is why your compass reading was off, and I can't answer that because as I mentioned you weren't near anything that I'd expect could influence it. I did notice that you've apparently only flown this aircraft a few times and it could be that it just needs a good compass calibration, though I can't say for sure. But again, there's definitely nothing here to indicate that any sort of cross-system interference took place between your drone and that of your friend.

From your post it sounds like the VIDEO signals switched but not the CONTROL signals.  If your friend was recording video when the switch occured there would be a file on his iPad/Android that shows the switch.  I think the control logs or DAT files will show you both had control of your drones.  If the video fails and you do not have visual of the drone the best thing to do is to change to map view to fly the drone closer to you.

Thanks to everyone helping out. I think the simplest explanation is the correct one, that I had the yaw error (unknown reason why), then it jumped into ATTI mode, which isn't the best... Approximately at the same time, the videofeed of my friends drone got screwed up at the same time, but this was a separate issue. But I dont think he actually saw my feed. For all I know, there was some local magnetic interference, which screwed up his position as well, which meant his videofeed only looked different. After all, this was nighttime, which made it harder to both see the drone, and navigate off the display.

I have Not enough Flight time to be commenting on this topic
Question .?
is this why the recommend in the setting you use the second controller Mode (2)..?
.
.

Hi, to chime in, I have had the exact same error twice now. I am flying, all of a sudden, the video signal switches to another drone and my drone starts behaving erratically. After 10-30 seconds of sheer anxiety, I regain control.

The new CAPTCHA crap is the last straw.
Im tired of DJIs crap and am removing my posts and will be on my way.

WT9BIND Posted at 3-18 16:48
Hi, to chime in, I have had the exact same error twice now. I am flying, all of a sudden, the video signal switches to another drone and my drone starts behaving erratically. After 10-30 seconds of sheer anxiety, I regain control.

Begging questions that DJI likely won't answer.

How many unique IDs for pairing is possible?
Is it possible for two different Drones & RCs to select (or have) same ID for pairing?
Is assigned unique ID for a drone used for all transmissions, including video feed?
Will paired Drone and RC reject all transmissions from other Drones and RCs?