CAUTION - DJI NAZA V2 GPS Unexpected Take-Off Hazard – Mike

Re: CAUTION - DJI NAZA V2 GPS Unexpected Take-Off Hazard – M

gmtx said:
So with Case X - loss of TX - the Phantom enters failsafe and returns home. But where is "home"?

One of the things I thought was interesting in the video I referenced was that the user said he thought his Phantom was trying to return to a previous "home" position from the last time he flew it. Is this possible - does the Phantom store home position in a non-volatile way that survives a battery pull/reboot? And what happens in Case X if you power on indoors and never get a GPS lock, hence no "home" position? No RTH, right? Would having S1 in ATTI prevent RTH from happening, even with GPS lock (or a "remembered" home positon)?

Sorry - I'm a newb with more questions than answers, but I really want to fly safely and unexpected (and potentially uncontrollable) takeoffs have me concerned.

There are a few factors here. First, I'm assuming failsafe is set to RTM and not just landing. I haven't tested yet, but "landing" mode shouldn't try to go up to 20m first, so it may or may not try to take off (see below).

If on RTH, and you are able to follow the standard best practice of locking in home point before arming motors, it will presumably take off, ascend to 20m, hover for 15 seconds and then descend back down in place.

If you DON'T follow best practices, and you arm the motors before 6 satellites are locked in, the Phantom won't have a home point set. I don't know what happens then. Some people say it heads towards some factory-initialized home point (read: the DJI factory in China). I'm not really sure how they figure that, or from a programming perspective why anyone would bother overriding the default of an initialized variable in such a useless way. It doesn't make sense.

As it turns out, turning off the TX while the Phantom is on the ground is a documented risk that DJI addresses (barely) in their failsafe documentation. They say don't do it, the craft WILL take off on its own. They don't say whether or not it'll happen if you're set to landing-only instead of RTH, but that's easy enough to test for anyone who wants to. So, at worst, they're aware of the behavior and it is WAI. Whether that's a good idea is a matter for discussion I guess:

Nzmv2-get_control.png


It's good to be mindful of the possibilities, but as a pilot, do keep in mind that the use cases here are really quite fringe (not that some people haven't been bitten by them, of course). If you make sure your tx is properly calibrated in assistant, be mindful not to set subtrims in a way that cause your Phantom to drift while at centered sticks (or if that's your thing, turn off RAP entirely), ensure your TX has charged batteries before a flight, wait for home point lock before arming, only arm the motors with the intent to promptly lift off, and keep a respectful distance from an armed quad (either in the air or on the ground), it's highly, highly unlikely that you'll be bitten by something like this. And now, if one of these odd situations does happen, you'll be more prepared!
 
Re: CAUTION - DJI NAZA V2 GPS Unexpected Take-Off Hazard – M

"Columbus,

I just ran a full suite of tests, and can confirm that elevator at constant non-centered position DOES trigger RAP. I would suggest hooking your Phantom up to the Assistant and see if you are inadvertently bumping the aileron a bit (or it may be drifting) when propping E for the test."

Thank you ElGuano for verifying that.

I will check this up on NAZA assistant.
 

Recent Posts

Members online

Forum statistics

Threads
143,066
Messages
1,467,357
Members
104,935
Latest member
Pauos31