Welcome to our Community
Wanting to join the rest of our members? Feel free to sign up today.
Sign up

Customer support after autolanding

Phantogram

Active member
Joined
Mar 21, 2020
Messages
28
Reaction score
16
I lost my drone recently. Made two flights just after sunset in dry conditions, 5°C, wind about 7-8 m/s at 50m altitude, wind perpendicular to flight direction. First flight went perfect, started with 69% battery charge, flew 7-8 minutes over a river, returned home, landing on the road, about 20% left. I changed the battery (66% full) and took off for the second flight in the same direction, nothing abnormal so far.

About 30 seconds in flight, I got a message that the drone was autolanding, and briefly a warning regarding overheating.
Unfortunately, the drone landed itself out of reach in the reeds of a salt marsh, not recoverable during night time / upcoming flow. There was no logfile of that last flight on my smartphone, possibly due to crash in the reeds / short circuit / distance between drone and smartphone / ...

I contacted Parrot customer service, quick response, but alas: they claim they can't give warranty without the logfile, although I presented them all screenshots, location info, previous logfile, complete detailed report of both flights of that day etc. , while clearly this was a product failure.

Planning to recover the drone now, but must find a way to do this without jeopardizing my life :s
Does it make sense to recover the drone, even if it was submerged already that first night during flow time?

Whatever the outcome, lessons learned: I'll always fly with a screen recorder from now on, to tackle the 'missing logfile' issue...
 
  • Wow
Reactions: edwardfcy1970
The "missing logfile" is, in my eyes, not a issue, but a feature.
Works well for Parrot, as soon a bird gets lost.
Because that logfile is only written AFTER the landing to your phone.
Which lets me think, that is wasn't only an autolanding, but also a fall, the last meters, shut off midair.

In case you can recover it, hopefully the SD card will have survived, in the FDR folder of that is the logfile of the last (all) flight (with that SD card)!

Good luck, but don't hold your breath!
 
You can retrieve the FF6 telemetry for the flight in question (if using on an Android phone).
 
Yes, using an android. Can you explain how to do it? Last logged flight in freeflight 6 is the one before the autolanding.

Possible that the drone tilted before actual landing, it landed here (possibly in reeds)

Screenshot_20210114-205414_FreeFlight 6.jpg
 
Last edited:
If you check in (from the root of the phone drive when you connect to the computer).

/ Android / data / com.parrot.freeflight6 / files / logs

Hopefully there will be a long list of files in there, using the file creation you should be able to locate the one for that flight in question. The filenames are a random UID, so will not relate to anything.

They are a compressed json file - decompress and use a json parser to format it nicely.

Sadly there are no useable readers yet for this format, so you will have to parse it manually to determine cause of autoland.

I suspect you will find that battery voltage dipped below ~6800mV briefly causing the autoland to trigger.
 
You made my day - at least: I found some files on the specified location, corresponding to the flight times.
i'll start organizing them in a readable format later today. Might be some task, content looks like this (see lowest pic)



Knipsel.GIF

Knipsel2.GIF
 
Like I said it is a compressed file (gzip), so decompress using a ZIP program such as winRAR, Winzip, 7zip, etc.
 
Quickly unzipped it during lunch time. Found a logging regarding motor error, I'll try to find out more later this evening.
Tnx for the support!
 
Motor Error entry will not mean what you think it does - that telemetry logs changes to status, the motor error entry is most likely entry to show no error.

It is a very hard log file to read as it relies on status codes and on-change logging even if the change is to say nothing wrong.
 
Do you know what the status numbers mean? There's indeed a status motor_error 0 in the beginning of each logfile. But in this one, there's also a status motor_error 2. I don't find that in other logs.
 
If the locally stored files really meant something, no offence, then people would have been able to get a warranty claim through them.

Basically there are two scenarios in which Parrot will help you:

1) the drone crashes and you retrieve the unit and can power it on; you will have the log file indicating power loss or whatever
2) the drone crashes and you have a screen recording showing you took off at 100% and didn't crash it into anything or flew kilometres away and then losing it or whatever

only in the second scenario will they help you without the actual drone
 
Do you know what the status numbers mean? There's indeed a status motor_error 0 in the beginning of each logfile. But in this one, there's also a status motor_error 2. I don't find that in other logs.

No I have not bothered no time to do so.

To fully decode all the status codes you will need to decompile the FF6 App and go through the code to translate status code to description (not hard to do but will take time).
 
If the locally stored files really meant something, no offence, then people would have been able to get a warranty claim through them.

Basically there are two scenarios in which Parrot will help you:

1) the drone crashes and you retrieve the unit and can power it on; you will have the log file indicating power loss or whatever
2) the drone crashes and you have a screen recording showing you took off at 100% and didn't crash it into anything or flew kilometres away and then losing it or whatever

only in the second scenario will they help you without the actual drone

Actually I think a member here who had dialog with Parrot stated that they have other logging data, which is this telemetry data as it is uploaded to Parrot by FF6 and forms part of the flight logs recorded.

So even if the final flight log is not on the App, Parrot do have a copy of the telemetry data for the final flight - and if the drone is under warranty they should analyse this telemetry data to determine cause of the crash.
 
No I have not bothered no time to do so.

To fully decode all the status codes you will need to decompile the FF6 App and go through the code to translate status code to description (not hard to do but will take time).
Ok. First thing to do for me now is to take the smartphone log of the last ok flight and compare it to the related json to see if it makes sense (ref reply of test2000). After that, pile through the log of the last flight and search for differences.
 
Actually I think a member here who had dialog with Parrot stated that they have other logging data, which is this telemetry data as it is uploaded to Parrot by FF6 and forms part of the flight logs recorded.

So even if the final flight log is not on the App, Parrot do have a copy of the telemetry data for the final flight - and if the drone is under warranty they should analyse this telemetry data to determine cause of the crash.

never ever has this happened, at least not a single instance of parrot providing warranty through a phone log (without the drone or screen recording)has ever been recorded on a forum or on the entire internet

edit: I meant that I only read claims to the contrary. Or is that people who didn't want to admit pilot error with denied claims?
 
Last edited by a moderator:
never ever has this happened, at least not a single instance of parrot providing warranty through a phone log (without the drone or screen recording)has ever been recorded on a forum or on the entire internet
And if you can combine screen recording with this logfile? Might be an option in the future.
 
And if you can combine screen recording with this logfile? Might be an option in the future.
Screen recording is enough; from the log file they can see the battery serial number I think. Anyway I hope you can retrieve the unit or file a claim (and get it approved)!
 
Whatever the outcome, I analysed the logfile to see what went wrong.
First of all, I arranged all logs versus timestamp (both status logs as error logs).

It appeard that the flight started normal.
After 18 seconds, there was a first product battery warning ("61"). The product battery log value however is "6951" just before that notice, and "6747" just after.

After 20 seconds, the bird got his GPS lock (before that, long / latt / alt are logged as "500" - after that time, it got actual values)

After 23 seconds, there was a second product battery warning ("60"). The product battery log value is "6956" just before that notice, and "6623" just after.

After 30 seconds, there was a 3rd product battery warning ("59"). The product battery log value is "6726" just before that notice, and "6655" just after.

After 31.3 seconds, there was a "forced landing error value 1" (at startup, this error value was set to 0)

Immediately after that, the bird started his landing. Altitude goes down. You see that I try to overrule using my sticks, no reaction.
Also 'MPP button value 16' is reported a lot, I think that's the return to home button on the control which I pressed multiple times.

Other error logs at same time:

31,295​
product_forced_landing​
1​
31,453​
product_alert​
6​
31,453​
product_fp_state​
1​
31,473​
product_flying_state​
4​
31,474​
product_rth_state​
2​
31,474​
product_followme_state​
0​


Bird descends / starts autolanding.

At 57.4 seconds, there's a second "forced landing" error, together with a motor error:
57,405​
product_forced_landing​
1​
57,563​
product_alert​
6​
57,563​
product_alert​
2​
57,563​
product_fp_state​
1​
57,563​
product_motor_error​
2​
57,563​
product_followme_state​
0​
57,564​
product_flying_state​
5​
57,575​
product_followme_state​
0​

At that time, the drone height is logged as 2,318m. It probably hit the top of the reeds, because the next logging, 0.2 seconds later, indicates height 0 and all speeds 0 (the seconds before, the z speed was -1,57m/s ).
The product roll was avg -0.18 in the seconds before, and was suddenly -1.2. The minimumvalue during the entire flight (besides of this event) was -0.26. So it probably rolled over touching the reed and crashed the last 2 meters.

What caused the event?
As far as I can see, a battery voltage drop.
The overheat error that I briefly saw, was probably at the time of touching the reed, and not the cause of the autolanding.

i also added the logfiles in attachment in excel and openoffice format. Anyone who feels to do some extra analysis, be my guest :)

(Error at 31.3 in the graph.)

flight graph.GIF
 

Attachments

  • 761d3b91-a2fd-451d-8625-0ea2868acef5~excel2000.zip
    42.4 KB · Views: 6
  • 761d3b91-a2fd-451d-8625-0ea2868acef5~openoffice.zip
    1.1 MB · Views: 3
  • Like
Reactions: Deleted member 3201
Excellent post. However, this shows that Parrot did have the ability to investigate crashes where they would be able to deduct if it was pilot error.

I believe with the firmware any 'high' level of voltage drop will force autolanding as the power would shut down.

I think that the motor error meant contact with the propellors, so that would be the crash
 
Indeed, but it' only thanks to this forum that I found out where to find the logfiles. Customer support never explained this to me.
I also send them this post in a mail, hoping that they will re-open the case.

And regarding pilot errors: yes they happen. Also to me. This crashed bird is my second Anafi. The first one was drowned when I hit a tree when flying over a river, only 3 weeks after I bought it. Never ever even thought to claim that one, just bought a new one, because it was my own fault.

But this one... clearly no pilot error.

I also understand Parrot, because they want to avoid unjustified claims. But in that case they should inform people that there are logfiles on the smartphone that can be used if the flight was not completed / not logged in the FreeFlight6 app ..
 
Last edited:
  • Like
Reactions: Deleted member 3201

New Posts

Members online

No members online now.

Forum statistics

Threads
5,294
Messages
45,048
Members
7,964
Latest member
edamame