That was scary

[DanMan32]

I just had a scary moment. I was flying up Weeki Wachee river from a Rogers Park. I reached a point at the river where it forks and I lost signal. I had been at this point about 10 minutes earlier and forked north which circles back. Instead of the P3A flying home on loss if signal, it apparently stayed there for 4 minutes until it reached a low battery RTH and just about made it back. Landed and ended flight (cut motors) just as it reached 10%.
Unless it had reset since this scary flight, I checked failsafe RTH setting and verified it is set to Return to Home, so it should have started coming home when it lost signal.

Here's the healthydrones analysis of the flight, but since signal was lost, the tablet won't show what the AC was thinking. HealthyDrones.com - Innovative flight data analysis that matters
I have the DAT files extracted from the AC if there's a way to analyze that.

I'm sure glad it made it back, with low battery as a backup. I had the flight in record, and the entire flight was recorded including the hover for 4 minutes it lost signal. I thought perhaps it had crashed during failsafe RTH before signal could be restored, or that if it was stuck, it would land where it was, which at that point I believe was over the river.

What I had found strange was that at the point I lost signal, I had been in that spot earlier in the flight with a decent signal if I remember correctly, which you can see in the flight log. I thought perhaps I had been at a lower height the 2nd time up the river but I was at the same height so that doesn't explain the different signal level.

I did get some nice recordings over there this afternoon. If I ever get around to editing them I'll post them.
I peaked the curiosity of some young kids and a dad I spoke with is considering getting a P3.

 

[GrammatonxXXxCleric]

You may have still had control signal. Video feed can cut out and you;ll still be able to move the drone although you have no idea of what it's doing. I just push left stick forward to gain altitude. A lot of the time that will re-establish video feed.

 

[GrammatonxXXxCleric]

RTH won't engage unless there is full signal loss. RTH initiated prbly due to battery failsafe .

 

[Oso]

You may have still had control signal. Video feed can cut out and you;ll still be able to move the drone although you have no idea of what it's doing. I just push left stick forward to gain altitude. A lot of the time that will re-establish video feed.

I agree. My immediate first thought after reading a couple of sentences was that this is a classic example of losing the downlink feed but still having control of the AC so RTH would not initiate.

 

[GrammatonxXXxCleric]

Yes, one thing you could do in this situation (Although... scary) is turn controller off. This will initiate RTH. But I'd try climbing up first. No lateral movements, just UP.

 

[DanMan32]

Well I just downloaded CsvView and analyzed the DAT file. Despite the tablet and flight log showing RC connection loss, the AC never considered a signal loss with the RC. I suppose I could have kept on flying, perhaps regain video and telemetry, then eventually it actually would be out of range. The RC did show a red light during those 4 minutes though.

 

[DanMan32]

If I lose video and telemetry for more than 10 seconds or so, I'll definitely try increasing height even if evidence suggests I have no control. And as Grammaton suggested, turn off RC.
The only problem with turning off RC is that it is a pain in the neck to get the tablet to connect again. When first starting out, I either have to reboot the tablet or power cycle the RC a few times.
This will also make sure that the AC does still detect signal loss and not a bug in the firmware.
I do someday want to test response time start to finish from signal loss to getting back in range. I'll do that in a park with ballfields where I can have the AC a decent distance away but still close enough and safe position to land if it had to. Then see how long it takes to react, ascend, turn, and start heading back.
I've experienced signal loss before where it seemed to take a long time to get back in range. After this experience, I bet the AC was still getting control signal from the RC, even if the RC was not getting telemetry/video, which prolonged actual RTH. You'd think that if the RC was not getting anything back from the AC, after say 10 seconds it would send a signal "I can't hear you" just in case the AC still thinks there's a connection.

 

[DanMan32]

I hope nobody thought I was spying on them, hovering out there like that for 4 minutes.

 

[GrammatonxXXxCleric]

I have noticed that DJI go loses contact WAAY more often than Litchi does. I use Litchi exclusively, but I like its VPS function when ultra low flights are needed. But losing video/rc is NOT acceptable and this is why I use Litchi. Also planning waypoints in the field is a major plus...

 

[GrammatonxXXxCleric]

I hope nobody thought I was spying on them, hovering out there like that for 4 minutes.

If you're low enough, people can use a baseball bat on it like a pinata.....