If you don't have enough battery to descend slowly from a high altitude, how reliable is it to just cut the motors and turn it back on at like 50-70m? What are the chances of recovery from the free fall? Anyone tried this before in low battery conditions?

There's a video of someone who successfully tried that, but I think he was lucky. I guess if you are likely to crash anyway it would be worth a shot -- the main question is if the Mavic can right itself after powering back on if it is upside down during free-fall.

Ex Machina Posted at 2017-4-21 10:04
There's a video of someone who successfully tried that, but I think he was lucky. I guess if you are likely to crash anyway it would be worth a shot -- the main question is if the Mavic can right itself after powering back on if it is upside down during free-fall.

Yeah that's precisely what I'm concerned about. If it fails to do that then it'll basically propel itself even faster to the ground.

I've seen this done a few times, but only on older models.

If you are up high, you should never be in a situation where you are low on battery.

DJI-Ken Posted at 2017-4-21 10:33
If you are up high, you should never be in a situation where you are low on battery.

I'm setting this in an unfortunate situation when you are actually running out of battery. Regardless of what DJI recommends to be "safe flying practices", such situations are almost inevitable. Battery failures such as the case with some of the Inspire 1s (suddenly dropping from 80% to <10%) cannot be completely avoided. Someone is eventually going to run into this situation.

What we want to know now is whether or not the drone's algorithms are sophisticated enough to recover it from such a drop with a worst case scenario of being upside-down.

It's definitely not something you should do expecting for your drone to survive.  I would not call it an evasive maneuver either.  I've only seen it done to find out what happens, but we were fully expecting it to slam down in the forrest.

method007 Posted at 2017-4-21 10:43
It's definitely not something you should do expecting for your drone to survive.  I would not call it an evasive maneuver either.  I've only seen it done to find out what happens, but we were fully expecting it to slam down in the forrest.

It definitely isn't.

However, I would like to know whether those cases when the drone did recover were results of DJI's software or were they just merely lucky.

There's a difference. Because even some people flew it into trees and the Mavic recovered just inches before it hits the ground. I suspect that the software is to an extent made to handle such situations. I'm just curious of the extent of its reliability.

Levurmion Posted at 2017-4-21 10:53
It definitely isn't.

However, I would like to know whether those cases when the drone did recover were results of DJI's software or were they just merely lucky.

Ive seen some AMAZING recoveries by the Mavic both first hand and on the internet. I have personally clipped a branch before and watched my Mavic do a barrel roll and once upright again the aircraft righted itself and hovered in place; blew my mind. Ive also seen people do exactly what your talking about..

The Mavic knows when its upside down so its  not going to power the motors if the aircraft is upside down I think it would actually wait until the aircraft is right side up and then apply the correct power to each motor to maintain stability. This has been tested a lot of you watch people who have crashed their mavic purposely for testing purposes. Its not something that is guaranteed and the maneuver would definitely have some inherent risk but I think its something that could be done multiple times with success.

Heres the video:

Heres a relating thread: http://mavicpilots.com/threads/m ... -real-flight.13335/

I do belive that once below a certain percentage one the motors are shut off they cannot be restarted