Automatic Landing Flyaway

[sar104]

I don’t doubt his assessment but he says final gps data point is not valid. How would I find final height? That’s a good idea.

The barometric data from after the reconnect indicate that it was at ground level.

 

[RayOZ]

The barometric data from after the reconnect indicate that it was at ground level.

I guess it’s on the ground then.

 

[sar104]

I guess it’s on the ground then.

That's where I would be looking.

 

[Buddhamcbuddha]

Thanks ver much for your analysis. I don't understand why there is no data to show that I steered the AC during Automatic landing. I am sure that I steered it away from the trees towards the field. During this procedure I lost connection. I assumed I lost connection due to the trees blocking signal due to the reduction in altitude. If it was nearby it should have been easy to find. Could it have flown away after losing connection?

 

[Buddhamcbuddha]

The barometric data from after the reconnect indicate that it was at ground level.

Here is the final altitude data I can see. Am I missing something?

g 18m 25s 354.6 ft 2,295 ft 1001.1​ 0%​ h 18m 30s 302.9 ft 2,955 ft 14.6​ 100%​ i 22m 47s 177.8 ft 37,818 ft 2.6​ 100%​ j 22m 59s -16.7 ft 38,055 ft 61.4​ 59%​

 

[sar104]

Thanks ver much for your analysis. I don't understand why there is no data to show that I steered the AC during Automatic landing. I am sure that I steered it away from the trees towards the field. During this procedure I lost connection. I assumed I lost connection due to the trees blocking signal due to the reduction in altitude. If it was nearby it should have been easy to find. Could it have flown away after losing connection?

I understand that you recall operating the sticks, and before it began to descend you had been applying some forward elevator to move it in the direction that you described, but that then stopped, and in fact you applied down throttle at the end, which is what started the final descent before it transitioned to low-battery autoland. At the end of the flight it was not moving laterally and going nowhere except down. It did not fly away.

Here is the final altitude data I can see. Am I missing something?

g 18m 25s 354.6 ft 2,295 ft 1001.1​ 0%​ h 18m 30s 302.9 ft 2,955 ft 14.6​ 100%​ i 22m 47s 177.8 ft 37,818 ft 2.6​ 100%​ j 22m 59s -16.7 ft 38,055 ft 61.4​ 59%​

Probably, because you keep looking at incomplete data. The DAT file barometric altitude data show that it ended up around 4.5 meters lower than the takeoff point. The ground level where it was descending appears to be around 4 meters lower than the takeoff point - suggesting that it is on or near the ground.



I really can't help you much further with this. The data are unambiguous, but if you want to believe that it's out in that field, or that it flew off somewhere else, then fine. Have you actually checked the position indicated?

 

[Baldric99]

Found this to be a very useful discussion. Thanks to all those who gave advice to the OP.

 

[slup]

1... I don't understand why there is no data to show that I steered the AC during Automatic landing...
2... I am sure that I steered it away from the trees towards the field. During this procedure I lost connection.
3... Could it have flown away after losing connection?

1 & 2: First of all so had you disconnects between AC & RC in the end ... so if you commanded or not during that period will not show in the logs. But when the connection just in the end reestablish you only had speed straight down & the raw GPS data shows a position just in the edge of that forrest ... so what you may or may not do before that, it all ended up there. And as I stated earlier the gimbal pitch wasn't straight down witch may have led you believe that you had passed the trees.

3: No ... as seen from both the TXT & DAT log the same last valid GPS position is the same in both logs, you have got that position so my advice is that you start your search there ... the bogus position out in the sea isn't valid, the navhealth was zero by then. Furthermore pure logic tells us that the M2Z couldn't have flown that distance of above 10 kilometers away with a battery at 13%, with autolanding in progress & in a time span that needed a speed over 40 m/s.

Here is the final altitude data I can see. Am I missing something?

g 18m 25s 354.6 ft 2,295 ft 1001.1​ 0%​ h 18m 30s 302.9 ft 2,955 ft 14.6​ 100%​ i 22m 47s 177.8 ft 37,818 ft 2.6​ 100%​ j 22m 59s -16.7 ft 38,055 ft 61.4​ 59%​

The basic is that the TXT log is recording from motors on to off & the DAT log from power on to off ... this if the RC/mobile device is in contact with the AC.

The TXT log ends at 1136sec after motor start, at that time your M2Z was in autolanding with only down speed at 37m above your HP ... then the log ends with a AC/RC disconnect and then the motors probably stopped hitting the trees ... so end of TXT log.

The DAT log though, still were recording as you luckily regained AC/RC connection ... when that log continues it shows the same position as the TXT log, no speed what so ever ... but a height of approx 4m below your HP indicating ground level. The DAT log records 851sec longer then the TXT log, indicating power still is on but nothing happens, all still in the same position (some small stick movements are registered, but AC is on ground with motors off so nothing happens), then that log also stops, probably due to AC/RC disconnection.

 

[Buddhamcbuddha]

I understand that you recall operating the sticks, and before it began to descend you had been applying some forward elevator to move it in the direction that you described, but that then stopped, and in fact you applied down throttle at the end, which is what started the final descent before it transitioned to low-battery autoland. At the end of the flight it was not moving laterally and going nowhere except down. It did not fly away.

View attachment 96640



Probably, because you keep looking at incomplete data. The DAT file barometric altitude data show that it ended up around 4.5 meters lower than the takeoff point. The ground level where it was descending appears to be around 4 meters lower than the takeoff point - suggesting that it is on or near the ground.

View attachment 96641

I really can't help you much further with this. The data are unambiguous, but if you want to believe that it's out in that field, or that it flew off somewhere else, then fine. Have you actually checked the position indicated?

Thanks for this.

of course I checked the area beneath the trees and in the trees themselves. The forest floor is a mossy/ grassy surface. The AC would be easily spotted. That is the only reason I have to consider other scenarios. It is a mystery for it to land in the forest or in the field and not be easily located.

 

[Prismatic]

Thanks for this.

of course I checked the area beneath the trees and in the trees themselves. The forest floor is a mossy/ grassy surface. The AC would be easily spotted. That is the only reason I have to consider other scenarios. It is a mystery for it to land in the forest or in the field and not be easily located.

Is it possible that someone else could have picked it up? The data is unequivocal, and you say the drone should be easy to find on the ground in that location.
Drones don't just vanish, and there are few possibilities considering what's known.

 

[Buddhamcbuddha]

Is it possible that someone else could have picked it up? The data is unequivocal, and you say the drone should be easy to find on the ground in that location.

No, I don't think so. It it private ground. Fenced in, with no walking paths passing the area. I did'nt see anyone and was within sight of the area within a couple of minutes of losing connection.

 

[sar104]

1 & 2: First of all so had you disconnects between AC & RC in the end ... so if you commanded or not during that period will not show in the logs. But when the connection just in the end reestablish you only had speed straight down & the raw GPS data shows a position just in the edge of that forrest ... so what you may or may not do before that, it all ended up there. And as I stated earlier the gimbal pitch wasn't straight down witch may have led you believe that you had passed the trees.

3: No ... as seen from both the TXT & DAT log the same last valid GPS position is the same in both logs, you have got that position so my advice is that you start your search there ... the bogus position out in the sea isn't valid, the navhealth was zero by then. Furthermore pure logic tells us that the M2Z couldn't have flown that distance of above 10 kilometers away with a battery at 13%, with autolanding in progress & in a time span that needed a speed over 40 m/s.




The basic is that the TXT log is recording from motors on to off & the DAT log from power on to off ... this if the RC/mobile device is in contact with the AC.

The TXT log ends at 1136sec after motor start, at that time your M2Z was in autolanding with only down speed at 37m above your HP ... then the log ends with a AC/RC disconnect and then the motors probably stopped hitting the trees ... so end of TXT log.

The DAT log though, still were recording as you luckily regained AC/RC connection ... when that log continues it shows the same position as the TXT log, no speed what so ever ... but a height of approx 4m below your HP indicating ground level. The DAT log records 851sec longer then the TXT log, indicating power still is on but nothing happens, all still in the same position (some small stick movements are registered, but AC is on ground with motors off so nothing happens), then that log also stops, probably due to AC/RC disconnection.

Did you notice that the offset time and tick data in the DAT file contain errors?

 

[sar104]

Thanks for this.

of course I checked the area beneath the trees and in the trees themselves. The forest floor is a mossy/ grassy surface. The AC would be easily spotted. That is the only reason I have to consider other scenarios. It is a mystery for it to land in the forest or in the field and not be easily located.

How accurate was your positioning when you searched?

 

[Prismatic]

And if you can't find it, several of my favorite distilleries are right there, so consolation beckons.

No better solace than a single malt!

 

[Buddhamcbuddha]

1 & 2: First of all so had you disconnects between AC & RC in the end ... so if you commanded or not during that period will not show in the logs. But when the connection just in the end reestablish you only had speed straight down & the raw GPS data shows a position just in the edge of that forrest ... so what you may or may not do before that, it all ended up there. And as I stated earlier the gimbal pitch wasn't straight down witch may have led you believe that you had passed the trees.

3: No ... as seen from both the TXT & DAT log the same last valid GPS position is the same in both logs, you have got that position so my advice is that you start your search there ... the bogus position out in the sea isn't valid, the navhealth was zero by then. Furthermore pure logic tells us that the M2Z couldn't have flown that distance of above 10 kilometers away with a battery at 13%, with autolanding in progress & in a time span that needed a speed over 40 m/s.




The basic is that the TXT log is recording from motors on to off & the DAT log from power on to off ... this if the RC/mobile device is in contact with the AC.

The TXT log ends at 1136sec after motor start, at that time your M2Z was in autolanding with only down speed at 37m above your HP ... then the log ends with a AC/RC disconnect and then the motors probably stopped hitting the trees ... so end of TXT log.

The DAT log though, still were recording as you luckily regained AC/RC connection ... when that log continues it shows the same position as the TXT log, no speed what so ever ... but a height of approx 4m below your HP indicating ground level. The DAT log records 851sec longer then the TXT log, indicating power still is on but nothing happens, all still in the same position (some small stick movements are registered, but AC is on ground with motors off so nothing happens), then that log also stops, probably due to AC/RC disconnection.

That is really amazing info. Thanks very much. I will go back and have another look in the last valid GPS position. I did search that area thouroughly but I must have missed it somehow. Thanks very much for all you help and great knowledge.

 

[Buddhamcbuddha]

How accurate was your positioning when you searched?

Yeah the last valid GPS position was quite easy to pinpoint as it was on the edge of the wood next to a solitary tree on the track that runs along the boundary.

 

[Prismatic]

.... I did search that area thouroughly but I must have missed it somehow. Thanks very much for all you help and great knowledge.

It's not uncommon for people to apply a high-visibility skin to their drone, for exactly this eventuality.
Too late, of course. ?

 

[Buddhamcbuddha]

It's not uncommon for people to apply a high-visibility skin to their drone, for exactly this eventuality.
Too late, of course. ?

Yup. We live and learn!

 

[Buddhamcbuddha]

Did you notice that the offset time and tick data in the DAT file contain errors?

View attachment 96654

No idea what this means I'm afraid.

 

[sar104]

No idea what this means I'm afraid.

That was just an observation for @slup - if you look at the DAT heights that I showed above in post #46 you will notice that the height goes to ≈ 0 well before the end of the txt log time. Fixing the DAT time data gives the correct view.



I'm not sure why the DAT file has data missing during periods when the txt log was correctly recording.