Really, thanks for your warm welcome, bro.
Plus for your general contribution to this topic, you've been really helpful.
I won't even make the effort to report a drawing of me pictured as a horse beaten to death, it's suggestive enough.
Fly-away and crash at 30mph...
- Thread starter Fracapouille
- Start date
Really, thanks for your warm welcome, bro.
Plus for your general contribution to this topic, you've been really helpful.
I won't even make the effort to report a drawing of me pictured as a horse beaten to death, it's suggestive enough.
From B - C the mean track is approximately 36° west of north, with the aircraft pointing 70° east of north. Forward elevator is therefore a thrust vector 106° from the track, which is close to orthogonal. It's a larger angle to begin with, hence the initial deceleration from B and smaller near C, hence the acceleration. So yes, expected.
You have all the analysis tools available to you that I have. If you want to make the case to DJI that some of its behavior in ATTI seems to suggest wind speeds that you don't recall then go right ahead. But remember that in ATTI there are only two effects that will change the direction and speed of the aircraft: pitch/roll and wind. If you look at the pitch and roll data you will see that it is not causing some of the velocity changes, therefore it must be wind or air currents.
Anyway, since you now seem to be determined to validate your hypothesis of equipment fault rather than simply to evaluate the data I'm going to leave you to it. This is no longer an interesting topic of discussion and I'm not going to waste any more time providing unwelcome analysis.
Meta4
Premium Pilot
- Joined
- Jan 8, 2014
- Messages
- 15,301
- Reaction score
- 9,243
There is a saying in English ... To flog a dead horse.
It means: to keep talking about a subject that has already been discussed or decided
Or ... To not know when to give up.
Ok, sorry about that, I didn't know this expression.
I overreacted, and tend to lose my sense of humor when I feel cornered.
I appologize.
This is not what I meant. To stay with the image of the vector, you won't consider the length of said vector. It's not like I'm just pushing softly, I'm 3s full throttle.
When starting this input (B), the track heading is 287.2°. So it's 143° (70+360-287) angle between trajectory heading and input direction. It looks closer to me of 180° than 90°.
I agree it is orthogonal at the end (340° roughly).
I just seems to me that there is a disproportion between the force of the input and the effect on the track. I just wanted you to take into consideration this force parameter.
As I said before, I don't and didn't expect much from DJI. I couldn't prove anything anyway. I was disappointed by their answer, but that's all.
Ok, that's a good point, thanks for pointing that out; that's what I missed, I didn't look closely enough to pitch and roll data. I'll try to figure that out from your graphs, and estimate wind effect on the flight. As you said, it seems consistent with the inputs though...
Come on, I never wanted to validate my hypothesis, it has always remained so to me. I just feel very frustrated that no one would accept this hypothesis,... The "drone failure" hypothesis has been discarded by the community from the beginning, leaving just the "pilot fault/wind" one. I said it before, I accept being at fault, I just can't accept an hypothesis would be rejected without a proper explanation as to why it can't hold together - and not only with a hastily prejudgment. You may think that you've already provided this explanation, it may be obvious to you (and fellow forumers) as -maybe- you've got a better overview of the data, and experience to read through these data, but, I'm sorry, it didn't seem that obvious to me.
Your analysis has always been welcomed, I truly appreciated it, and I never questioned it. I asked for explanations about your interpretation, and shared my doubts.
The "drone failure" hypothesis has not been discarded, but there are no data to support it. The flight looks like a poorly controlled ATTI flight. Without detailed wind field data, some of the flight behavior is not entirely obvious, but none of it is clearly due to anything other than either stick inputs or air currents that would be common in such a topography. There is no definitive proof from the log files, but Occam's Razor says it was pilot error.
Just to add a perspective, as a public safety UAS pilot, we wouldn't launch a P4P (which we fly) in a GPS denied/ degraded environment. We require a minimum of 8 locked satellites for takeoff and treat a loss of GPS as an in flight emergency "Land Immediately" situation. The reason for that is exactly this circumstance. Flying a manual photogrammetry mission is task loading enough, add in a difficult flight environment and 0 GPS hold and it can overwhelm even an experienced pilot. We fly in dual operator with our P4Ps and I still wouldn't fly without GPS.
I'm not implying the OP is wrong, professional flying is about Risk Management. Risk is always relative when it only involves loss of the drone as there didn't appear to be a risk to persons here. A loss of a single P4P wouldn't move the needle budget wise for my operation, but it might be different for you and that should be factored in the risk analysis.
As far as the flight data analysis, way over my head, but I enjoy learning about it and I appreciate the OP sharing this incident so we can all learn what we can. I just wanted to add an Aeronautical Decision Making perspective to the technical one.
Fly safe
I'm not implying the OP is wrong, professional flying is about Risk Management. Risk is always relative when it only involves loss of the drone as there didn't appear to be a risk to persons here. A loss of a single P4P wouldn't move the needle budget wise for my operation, but it might be different for you and that should be factored in the risk analysis.
As far as the flight data analysis, way over my head, but I enjoy learning about it and I appreciate the OP sharing this incident so we can all learn what we can. I just wanted to add an Aeronautical Decision Making perspective to the technical one.
Fly safe
Roger that.
Thanks for your interesting feedback.
The survey was done in a touristic place, and I demanded the intervention to take place while the site was closed to both tourists and workers. The only people allowed were land surveyors that were attending the flight and surveying GCPs, they were aware about what was going on and security measures. If security conditions were not that bad for people at least, what was wrongly assessed on my side, is that the time slot was very short: the site is open 7/7, impossible to close, and I had to fly before opening hours, but with enough luminosity. The site is not close from the office either... It lead to some kind of emergency to launch the mission, and this is something I have to get more aware of. Postponing a flight is not an easy decision to take, when you try to satisfy both the client and contractor, and especially when you don't know if you will be able to fly in better conditions (weather, GPS) the next day...
You're totally right regarding risk management. The loss of a P4P won't jeopardise my business, it is half of the yearly budget for eBee batteries and replacement parts... But this experience helped me learning a lot of things though - thanks to all of you - and it will help me to improve my working methods through risk management. I took the chance, I shouldn't have, but the risks were accepted. Next time, I don't know if I would take the chance again or not, but I'll be certainly far more aware about the risk-taking. On a longer perspective, it translates by a better judgment of maintenance (including replacement...) cost of such a device for photogrammetric purposes. Lessons learned...
Fracapouille, I meant no harm by the image, just an attempt at humor. Sorry you took it to mean anything else.
Similar threads
