Mavic Mini signal loss & failed RTH - help with analysis please

[Juz]

A couple of days ago I experienced my first significant UAV mishap and lost my Mavic Mini. I'm not looking to blame anything, just wanted to get some assistance with analysing the factors that might have contributed so that I can reduce the risk of it happening again.

I'm still a relatively inexperienced pilot. Had a Phantom 3 Pro around 6 years ago which I only flew for approx 5 hours before having to sell as I was migrating. Then I bought this Mavic Mini around a month ago, had 2.48 hours flight time from 19 flights. I consider myself moderate risk-taking with where and how I fly it, but have found the RTH always got me out of trouble when I flew beyond the signal range and have always then been able to reconnect and take back over control with no dramas.

This time however I had a sudden complete signal loss and disconnection, the RTH function didn't save me (more on that later), and I wasn't able to get close enough to resynchronise the connection before the battery would have died.

I wasn't able to find a Mobile Device DAT file, so only have TXT log unfortunately.
I have also uploaded the streamed video footage of the flight.

To view this content we will need your consent to set third party cookies.
For more detailed information, see our cookies page.

Accept third party cookies

When I watched back the flight log in the DJI Fly app Data Centre the first issue that stands out is that my RTH was set too low. I thought that I had this set to 160m, which in the hilly regions where I tend to fly is usually enough that if I lose signal behind a obstacle, increases the altitude by a few metres and quickly reconnects, allowing me to resume control and bring it back to safety. I usually check this before every flight, but on this occasion I missed that it was set lower.

When watching the video the next issue I noticed is that after I ascended the hill, in which I sat inside the 'trough', I then flew further behind the peak of the hill which was to the left of the earlier flight path. I suspect that this is what caused my sudden loss of signal. At this stage I was at 134m altitude from take-off at sea level, and due to RTH being set to 100m, I think if it did begin a RTH process after I lost signal, would likely have caused it to crash into trees/rocks/the ground - would this be safe to assume, or does it have any more capable collision detection than this?

The last things I wondered about were other factors such as whether the wind or possible bird attack might have been involved. It was quite a windy day, but to be fair most of the flights I've done with this Mavic Mini have been similar conditions and I had been really impressed at how well it dealt with strong wind considering it's diminutive size - so there wasn't anything extraordinary about the wind on this day that made me think it wouldn't have been able to remain stable and comfortably fly against it. Finally, when I went to try and look for the missing device I noticed a couple of Brahminy Kites circling around the area, though these aren't typically huge birds, I wondered whether one might have attacked it. When I watch back the log and video though, it seems my signal suddenly dropped off very quickly, and the video footage at this time doesn't seem to suggest it was struck by another flying object, so I don't think this is the case - but would welcome more expert opinions?

I would like to buy a replacement. I was initially planning to just get another Mavic Mini as it fulfils all of the features I want, and the financial impact of losing it is much less than most other models. But I am now considering whether had I bought a Mini 2 or Air 2 with OccuSync 2.0 in the first place, perhaps I wouldn't have lost it as the signal link might have been stronger. What do you think?

Thanks

 

[Thomas B]

If you lose VLOS of the drone you lose control. True of all DJI drones. Occusync X.X is not a factor. Don’t fly it out of VLOS if you want it back and there are intervening high points.

 

[Meta4]

the first issue that stands out is that my RTH was set too low. I thought that I had this set to 160m, which in the hilly regions where I tend to fly is usually enough that if I lose signal behind a obstacle, increases the altitude by a few metres and quickly reconnects, allowing me to resume control and bring it back to safety. I usually check this before every flight, but on this occasion I missed that it was set lower.

That's a very "interesting" part of the world.
Last time I was there, consumer drones hadn't been invented, but it would be a great place to fly and photograph.

Here's the detail of the far end of your flight.
The straight line is the RTH path from loss of signal.

You can see that there is a high point (near the red arrow) and you flew in to the north of it.
But it is on the direct line to the home point from where signal was lost - the RTH path.

When watching the video the next issue I noticed is that after I ascended the hill, in which I sat inside the 'trough', I then flew further behind the peak of the hill which was to the left of the earlier flight path. I suspect that this is what caused my sudden loss of signal. At this stage I was at 134m altitude from take-off at sea level, and due to RTH being set to 100m, I think if it did begin a RTH process after I lost signal, would likely have caused it to crash into trees/rocks/the ground - would this be safe to assume, or does it have any more capable collision detection than this?

When signal was lost the Mini was 442 ft higher than the launch point (already higher than the RTH height) so the RTH would have been at that altitude.

RTH would have taken the Mini directly toward the high point, only about 20 ft lower than the RTH path (but Google Earth heights are only approximate).
You also have to add the height of any trees and from memory the trees on those hills are over 20 ft high.


Obstacle avoidance has its limitations including small branches, so it can't be counted on when you need it.

It was quite a windy day, but to be fair most of the flights I've done with this Mavic Mini have been similar conditions and I had been really impressed at how well it dealt with strong wind.

You didn't leave the Mini hovering during the flight, which would have given me some clues about wind strength.
The Mini isn't very good at dealing with wind and the return flight would have been into the prevailing wind direction.

I noticed a couple of Brahminy Kites circling around the area, though these aren't typically huge birds, I wondered whether one might have attacked it.

Brahminy kites would be a very low risk for attacking a drone - they usually find their food on the shore or the sea surface and aren't usually aggressive.

the video footage at this time doesn't seem to suggest it was struck by another flying object, so I don't think this is the case - but would welcome more expert opinions?

The pitch and roll data were stable, indicating no collision or bird strike near the end of data.

I am now considering whether had I bought a Mini 2 or Air 2 with OccuSync 2.0 in the first place, perhaps I wouldn't have lost it as the signal link might have been stronger.

The mini 2 is much more capable in wind and the Occusync is many times better for maintaining signal.

That high point is the most likely cause of the failed RTH.
The last location data puts the Mini at -19.11108 146.83319.
If the Mini could make headway against the wind, it should be between that location and the high point.
It could have hit a tree and fallen or been stuck in branches.

 

[maelstrom]

The mini 2 is much more capable in wind and the Occusync is many times better for maintaining signal.

But if you fly behind a hill, as the OP did, you’ll lose the signal whether you have Occusync or not.

 

[tlswift58]

So, if you live close and this is not a horrible bushwhack - time to go back and see if you can find it. You never know - you may luck out.

 

[BenAir]

Let us know how you go with the search. People do find their drones. Makes me smile when they do

 

[Juz]

Thanks for the suggestions and particularly the useful detailed flight log analysis.

I was away on the last day of a holiday when this happened so unfortunately didn't have sufficient opportunity to try to retrieve it. I had already walked the full length of beach trying to resync the connection whilst hoping it still had battery life (leaving my heavily pregnant wife sat out in the hot sun with no way for us to stay in contact) - I did then begin to try climbing up the rocky hill for a few minutes but I was barefoot and realised even if I had shoes it would have taken me a couple of hours, so I had to abandon it...

With regards to VLOS, honestly I don't think most people would even find flying their drone remotely interesting if everybody kept it in close enough range to still see it with the naked eye. These things are so tiny that it's difficult to spot or hear in the sky even just 30m away.

Hopefully I will be back in the skies with a new Mavic soon

 

[Prismatic]

But if you fly behind a hill, as the OP did, you’ll lose the signal whether you have Occusync or not.

Not necessarily. I'm not an RF engineer, but I know RF reflection is a thing. If there are RF-reflective topographic features you might (no guarantees!) get away flying behind a hill.

For example, I once sent my MA1 out around (behind) a massive stone monolith in the desert (say 500m W x 300m H x 50m D). WiFi signals don't pass through sandstone, of course, but there were surrounding cliffs and slickrock that (apparently) kept the bird connected.

And while not the pilot, I once saw an M2p flown across a mountain valley (2km, say), then another 2km up a side canyon. After turning up that canyon the bird was definitely not VLOS and definitely well behind a ridge. The video feed and control link were stable throughout and the M2p came home fine.

I assume the even higher ridge on the far side of that canyon was RF-reflective, and the bounced signals were sufficient, but that's just a guess. The low vegetation there was buried under a 1 or 2 m blanket of snow, and I don't know how that affects RF reflection. I really felt the pilot was banking on blind luck, but it made me an Occusync believer.

 

[maelstrom]

If there are RF-reflective topographic features you might (no guarantees!) get away flying behind a hill.

Only if you're very lucky!

 

[JAW]

With regards to VLOS, honestly I don't think most people would even find flying their drone remotely interesting if everybody kept it in close enough range to still see it with the naked eye.

Not to be argumentative, but IMHO that is totally not true. Perhaps if you changed “most people” to ”many people” it would be more accurate.