fly away vs atti

[yorlik]

SO I've been reading hours on end all I can find on the fly away topic, wondering if there isn't a or multiple methods to prevent it or stop it once it happens.

So far, it seems to me all posts say 'there was nothing I could do to stop it flying off.' Some even stating that switching to atti mode made no difference. The most plausible reason I have found is that home position is set wrong, maybe even 10-15 miles away. So when RTH starts it heads out. But I cannot see why switching to atti mode would not stop this in its tracks?!? Atti mode is NON gps, just using compass and barameter. So if an errant home point is set, switching to atti mode MUST stop it.

Could it be that in the adrenaline flowing excitement of the fly off folks report trying atti but actually did not?

Could it be the errant home gps position is NOT the issue?

What am I missing?

 

[BlackTracer]

SO I've been reading hours on end all I can find on the fly away topic, wondering if there isn't a or multiple methods to prevent it or stop it once it happens.

So far, it seems to me all posts say 'there was nothing I could do to stop it flying off.' Some even stating that switching to atti mode made no difference. The most plausible reason I have found is that home position is set wrong, maybe even 10-15 miles away. So when RTH starts it heads out. But I cannot see why switching to atti mode would not stop this in its tracks?!? Atti mode is NON gps, just using compass and barameter. So if an errant home point is set, switching to atti mode MUST stop it.

Could it be that in the adrenaline flowing excitement of the fly off folks report trying atti but actually did not?

Could it be the errant home gps position is NOT the issue?

What am I missing?

I started this thread viewtopic.php?f=27&t=33213 a while ago which I think answers a good number of these errant flyaways, but very few seemed interested. Oh well. Tell me what you think.

 

[bbfpv]

When mine flew away, a few minutes prior I had gotten the "phantom connection broken" message. I immediately went to failsafe and she came back to me so I took control and redirected her back out again, same approximate location and got "phantom connection broken" again. Tried failsafe but this time it didn't work. Immediately switched to atti and tried to throttle up a few feet higher so I could get her away from some trees and the building I was shooting but she didn't respond, started drifting with the breeze, and that's the last I saw of her. And the "last seen" pin was exactly where I last saw her before she drifted away.

 

[yorlik]

When mine flew away, a few minutes prior I had gotten the "phantom connection broken" message. I immediately went to failsafe and she came back to me so I took control and redirected her back out again, same approximate location and got "phantom connection broken" again. Tried failsafe but this time it didn't work. Immediately switched to atti and tried to throttle up a few feet higher so I could get her away from some trees and the building I was shooting but she didn't respond, started drifting with the breeze, and that's the last I saw of her. And the "last seen" pin was exactly where I last saw her before she drifted away.

Ouch. As bad as it is, THIS at least makes sense! So second time it lost contact but not before changing to ATTI worked. So then since you were out of range, it just DRIFTED away on the wind - as expected unfortunately.... wonder if it maybe makes sense to set RTH height REAL high so it goes up and gets clear line of sight contact again to reconnect with lost controller sig .....

Guess the moral of your story is very informative too: If loose sig and RTH works, don't go back to same spot that lost sig to begin with. Corollary might also be to NOT switch to ATTI if near edge of communications or it may just drift away!

So many variables!

THANK YOU for your comments. So sorry about your dji. Why I am anxiously awaiting my rf beacon from NZ before doing any far flights....

 

[yorlik]

Does DJI follow these forums?

If it were MY product, I would have a person scouring forums on my product for new ideas to improve it. I have read so many now! Here is a perfect example of an EASY improvement:

DJI, IF ONE IS IN ATTI MODE AND CONTROL SIG IS LOST - SWITCH THE DAM THING TO GPS SO THERE IS AT LEAST A CHANCE IN HE11 OF RTH. IF AFTER 10 SEC IT IS NOT HEADING BACK IN CONTROL, LAND THE DAM THING STRAIGHT DOWN!

How the heck hard is thinking of this stuff????

 

[bbfpv]

No kidding. The crazy thing is you don't really comprehend how far one of these things can drift until you plot the approx speed * battery remaining radius on a map. Taking into account I know it didn't fly into the building or trees it was near, it ended up being a cone extending across my entire town. I have an "if found please call" sticker on it, but considering that in its projected flight cone there are 3 golf courses, and it snowed about 10" the following day, even if someone does find it in the spring, she's definitely trashed. What's even more annoying is my flytrex came later that week......

 

[landmannnn]

I know it has been said before, so apologies, but there is not one single reason for flyaway. Sadly this means that that there is no simple method of recovery, although atti might be one of them.

The most common causes seem to be novice pilot error, leaving S1 and S2 activated and not noticing, erroneous gps positon as start up, loss of gps signal, pushing the boundaries of height and distance, loose gps connector, software hiccups, possible interference and compass calibration. Nobody really knows because the evidence is lost.

If it happens to you there several things to do:
Don't panic!, check S1 and S2 both up, check your iOSD feeds (sats, distance etc), check antenna angle, check your lights if your are near enough, try to land in gps, try to land in att, try HL, try failsafe, turn off transmitter - wait and turn it on again. As above, don't go switching madly, wait to see what happens at each stage.
As a final resort kill the motors and crash.

If none of these work you could try running after it in the hope it recovers, ideally you have fitted a gps tracker.

By the way, I strongly believe that flying straight off the ground is a major factor, gps signals do odd things at ground level (multipath, absorb etc) and compasses are easy to upset. Sorry, prevention wasn't the post topic!

 

[yorlik]

Does DJI follow these forums?

If it were MY product, I would have a person scouring forums on my product for new ideas to improve it. I have read so many now! Here is a perfect example of an EASY improvement:

DJI, IF ONE IS IN ATTI MODE AND CONTROL SIG IS LOST - SWITCH THE DAM THING TO GPS SO THERE IS AT LEAST A CHANCE IN HE11 OF RTH. IF AFTER 10 SEC IT IS NOT HEADING BACK IN CONTROL, LAND THE DAM THING STRAIGHT DOWN!

How the heck hard is thinking of this stuff????

You should really read the manual before making that kind of suggestion. What you are suggesting is not an improvement. It's a feature that has been implemented in NAZA for years.

If you are flying in ATTI and lose control signal RTH is initiated.

Thanks Canard, I AM trying to learn as a newbie here!

Unfortunately, I DID read the manual - like 10 times. Still digesting it too!

Double unfortunately, your reply is wrong also. I just went and read it slowly yet again and it clearly states that if in non GPS mode, and failsafe starts, it will simply LAND from where it is. Which of course is all it can do in ATTI mode if you don't try to regain GPS control. So it does NOT initiate a RTH sequence. Unfortunately for bbfpv, when his was in atti and lost connection, instead of simply landing, it drifted away.

So I say my statement is still valid: if in ATTI mode and control is lost, it needs to either land as promised in the manual, or try switching to gps mode and attempt a RTH. Then a failsafe timer needs to be running to say if in 10 sec it is NOT doing a successful RTH sequence, LAND NOW. As a programmer, I can tell you this would NOT be hard to implement!

The more I learn, the more I wish DJI would open source their software so others could begin to tweak it better. They have such a fantastic product now and it will only get better with time. SO MANY good little software tweaks have been mentioned it is a shame they cannot implement at least some of them sooner than later.

 

[BlackTracer]

Does DJI follow these forums?

If it were MY product, I would have a person scouring forums on my product for new ideas to improve it. I have read so many now! Here is a perfect example of an EASY improvement:

DJI, IF ONE IS IN ATTI MODE AND CONTROL SIG IS LOST - SWITCH THE DAM THING TO GPS SO THERE IS AT LEAST A CHANCE IN HE11 OF RTH. IF AFTER 10 SEC IT IS NOT HEADING BACK IN CONTROL, LAND THE DAM THING STRAIGHT DOWN!

How the heck hard is thinking of this stuff????

You should really read the manual before making that kind of suggestion. What you are suggesting is not an improvement. It's a feature that has been implemented in NAZA for years.

If you are flying in ATTI and lose control signal RTH is initiated.

And additionally, if in atti with less than 6 sats and control signal is lost, it will land where it is, in atti mode, drifting with the wind on the way down.

 

[packetlos]

With my own experience of a fly away I don't think most of these are pilot error, especially if the phantom takes off in a random direction at high speed.

My theory is these are losing the RC control signal on the 5.7-5.8Ghz band and responding to interference instead, there is supposed to be a built in algorithm designed to detect improbable commands but I suspect it doesnt work all that well. If you are lucky the S1 switch commands will get through and you can get it in to Atti or even better fail-safe RTH and the RC control is removed from the equation, all the more reason to make sure the home point is set.

 

[TuT]

From the NAZA manual:
If you get out of radio range - "Attitude Mode: (1) the aircraft will level its attitude immediately (2) 3 seconds later, failsafe is triggered and aircraft will start to go home. (3) If signal is regained during (1) or (2), it will resume normal flight immediately."

 

[NRL777]

With my own experience of a fly away I don't think most of these are pilot error, especially if the phantom takes off in a random direction at high speed.

My theory is these are losing the RC control signal on the 5.7-5.8Ghz band and responding to interference instead, there is supposed to be a built in algorithm designed to detect improbable commands but I suspect it doesnt work all that well. If you are lucky the S1 switch commands will get through and you can get it in to Atti or even better fail-safe RTH and the RC control is removed from the equation, all the more reason to make sure the home point is set.

My flyaway and recover had something to do with another friend flying his 450 with his FPV on 5.8ghz/Fatshark..
As I was flying in ATTI mode, he had turn on his FPV transmitter/receiver and my vision+ went left... when I panicked and yelled flyaway, he had turned off the Fatshark and onboard transmitter on his (he was on the ground doing some tests) and I regained control.. about the only thing I had control when it flew away was height so as it got further and further the only thing I could do was to make it climb higher and higher.. you could bearly see it.. so I used RTH to make it fly back.

Not sure, has anyone confirmed or seen confirmed that 5.8Ghz video with the popular FPV frequencies don't jam the Vision 5.8?

 

[CyberCitizen]

I Had A Issue Where The Thing Flipped And Looking At The Flight Log The Speed And Everything Else Increased So There Are Some Issues There, However Had DJI Replace The MC.

 

[billn]

Double unfortunately, your reply is wrong also. I just went and read it slowly yet again and it clearly states that if in non GPS mode, and failsafe starts, it will simply LAND from where it is. Which of course is all it can do in ATTI mode if you don't try to regain GPS control. So it does NOT initiate a RTH sequence.

No. You're still wrong.

I agree with you.

What you have shown is from the NAZA-M manual and the Phantom manual says it a little differently. In the Phantom manual it says that if its

"4.2
Initiating the Failsafe mode from different flying statuses will result in different processes.
Ready to Fly(NON-GPS)-Automatic Landing."

It says flying statuses and NON-GPS status less than 6 sats, and it does not say anything about ATTI MODE (or Manual mode). The NAZA-M manual clearly says that in ATTI MODE it will RTH.

If the Phantom goes into failsafe because it has lost signal you can flip the switches all you want and it won't do a thing until it recovers signal.

I have had it happen, I was flying in ATTI mode when I got some interference and lost WiFi and control at the same time and the Phantom was well within sight when it happend. It started its Return to Home procedure and it regained the control signal and stopped coming home. I had the camera running at the time and you can clearly see when RTH started and ends just like the NAZA-M manual says.

 

[yorlik]

Boy there is a lot to learn! So much contradictory info! Thanks to all for correcting me. My apologizes Canard! I guess I should not have ASSumed that 'ready to fly (gps)' and 'ready to fly (no gps) in manual that came with my drone actually meant ATTI mode in the no gps statements since it clearly says no RTH just land where it is. For the first few flights I had, without sats it sure SEEMED to fly like it does now in ATTi too. Thank you all for posting links to better manuals that tell more accurate stories.

 

[dalebb]

Boy there is a lot to learn! So much contradictory info! Thanks to all for correcting me. My apologizes Canard! I guess I should not have ASSumed that 'ready to fly (gps)' and 'ready to fly (no gps) in manual that came with my drone actually meant ATTI mode in the no gps statements since it clearly says no RTH just land where it is. For the first few flights I had, without sats it sure SEEMED to fly like it does now in ATTi too. Thank you all for posting links to better manuals that tell more accurate stories.

I'm curious, I have a post going about GPS SATS, where do you live thats you don't get enough SATS?

 

[yorlik]

dayton, OH. I went thur thru sun last week with NO sat because of such heavy overhead clouds.

 

[yorlik]

glad to hear that!

someone needs to tell gps sats over my head that - they didn't get the memo.

and when it rains heavy, tell the direct TV sats that they are no longer allowed to shut my tv sig off for the duration.

and tell the same gps sats that make my car garmin stop working when really cloudy to stop acting like children!

 

[yorlik]

Canard, lets just agree to disagree.

I don't need to read your facts to tell my how gps works in my life. sats do go away in clouds, heavy snow, and heavy rain in my world. I am glad this is not so in yours.

I am old enough to know what I see and experience.

If you want to just disagree with everything I say, thats ok. I will just ignore it from now on.

Jeesh!

 

[yorlik]

If you want to just disagree with everything I say, thats ok.

I haven't disagreed with everything that you have posted. Just the stuff that was absolutely incorrect (RTH ATTI behavior, what's in the manual, GPS signals being affected by clouds, etc.).

You are certainly entitled to your opinions (however mistaken) but you aren't entitled to your own facts.

OK, I can't resist, so will reply one more time.

- RTH behavior as explained in dji manual as 'ready to fly no gps' I was correct and you were wrong.
- what's in the manual - I was correct, you are wrong. (my v3 drone did not come with any such naza manual you seem to refer to)
- gps effected by rain,snow,clouds, imperically I know from my experience you are wrong.

Enjoy your negative argumentative life my small friend.