I have a P4P+ running GO App 4.1.4 on Android OS and the App crashes far too often and it seems to happen most frequently when in intelligent flight modes - tap fly, POI, and Waypoint.  This is disconcerting because all situational awareness of what the aircraft is doing is lost, battery level, orientation, etc..  It is doubly troubling when in some intelligent flight modes because the joysticks no longer control the aircraft.  For example, in Waypoint mode you can't control the altitude while it is flying the mission and it continues to fly the mission if the GO app crashes.   Another example, if the app crashes in POI mode, there is no way to fly it home using controls, it will continue to circle until battery runs out.  So essentially you no longer have control of the aircraft.   With the P4P+ embedded display there is no way to reboot the GO app without turning the RC off and on, so there is no control at all during the reboot and even if it does eventually reconnect, and when GO app comes up sometimes it does not go into the intelligent flight mode it was in when it crashed, so you can't get it out of the flight mode.    This is a safety issue, if an aircraft were to appear you would be unable to take action to avoid it.  Pressing RTH is not an option because in that situation the last thing you want to do is have the aircraft climb to RTH altitude.  I have seen a lot of instances of the GO app crashing on this forum and others on standards RCs that use an android phone or tablet, so I know DJI is aware of it but nothing has been done.  The android GO App is still 5 versions behind the IOS GO app and its even worse for the P4P+ and P4A+ controllers.   I can only surmise that since there are many more customers using IOS devices, the android GO app is a lower priority and the embedded display version even lower still.


This situation is unacceptable, so this is what I am going to do and I urge others who are experiencing GO app crashes to do the same.

Every time the GO app crashes and I lose control is a safety issue, so I am going to report it on the FAA incident reporting system.  

Perhaps the FAA can get DJI's attention.

Most if not all automatic missions and flights can be stopped without using your app as long as you have your AC in VLOS which you should have particularly if your getting the FAA involved. First what you need to do is refer to your manual to find out how you stop all auto flights without app , because I don't have the time to explain all of them to you.
I will tell you that if your flying litchi you lose app just switch to sport and back to P and fly home, I'll give you this one because it's not in your manual.

PAUSE BUTTON WILL BECOME YOUR FRIEND!!!!!!!

The sky is falling. Let's get the government involved. I'm sure it will solve all of our problems.

I don't have a P4 so forgive me if this isn't possible, but can't you just press the RTH button, invoke RTH, and then press the RTH button again to cancel RTH?   Then you can fly it home and land it safely.

What happen if you change to P mod when you are using POI and the app crashes?

I‘m sorry if this bothers you. The drone is still under your control when the APP crashes. Take care and you can fly it back safely.

Mark The Droner Posted at 2017-9-19 17:42
I don't have a P4 so forgive me if this isn't possible, but can't you just press the RTH button, invoke RTH, and then press the RTH button again to cancel RTH?   Then you can fly it home and land it safely.

Yes you can that’s exactly how you cancel Rth.

I think some people, especially new folks, get so dependent on the mobile device and app that they don't think about the fundamentals - that is - the option to fly it old school like the P1 which never had a screen or an app.

DJI Susan Posted at 2017-9-19 19:12
I‘m sorry if this bothers you. The drone is still under your control when the APP crashes. Take care and you can fly it back safely.

That is correct only if you are not in an intelligent flight mode such as POI, where the joysticks are reprogrammed.  When my P4P+ is in POI and circling and the GO app crashes it continues in POI mode.  You can pause it, with the pause button, but there is no published method that I have seen for getting it out of POI so it can be flown home with the joysticks.   It has been suggested to briefly put it in "S"  mode but there is nothing published about whether that would erase the home point and RTH info.  Can you verify that it does not?  Also, is there some other trick to reboot the display such as holding display button down for 10 seconds or something?   Otherwise the only way to get the GO app back is to turn RC off and on.

wataero Posted at 2017-9-21 13:19
That is correct only if you are not in an intelligent flight mode such as POI, where the joysticks are reprogrammed.  When my P4P+ is in POI and circling and the GO app crashes it continues in POI mode.  You can pause it, with the pause button, but there is no published method that I have seen for getting it out of POI so it can be flown home with the joysticks.   It has been suggested to briefly put it in "S"  mode but there is nothing published about whether that would erase the home point and RTH info.  Can you verify that it does not?  Also, is there some other trick to reboot the display such as holding display button down for 10 seconds or something?   Otherwise the only way to get the GO app back is to turn RC off and on.

It has been suggested to briefly put it in "S"  mode but there is nothing published about whether that would erase the home point and RTH info.  Can you verify that it does not?

On the non+ models, it is commonly used to interrupt programmed flight in POI or Litchi so it's definitely worth trying.
The home point is stored in the Phantom so that it can RTH if disconnected.
Flicking the mode switch will not have any effect on the stored home point.

Labroides Posted at 2017-9-21 13:40
It has been suggested to briefly put it in "S"  mode but there is nothing published about whether that would erase the home point and RTH info.  Can you verify that it does not?

On the non+ models, it is commonly used to interrupt programmed flight in POI or Litchi so it's definitely worth trying.

Thanks, that is good to know.  

I find it amazing that DJI's flagship Phantom model seems to have these kinds of issues...

With respect to the FAA involvement, yeah nice thought, however not so much.  Rather be careful what you ask for as they are likely far more sympathetic to DJI than your issue, although your point has some merit.

wataero Posted at 2017-9-21 18:19
Thanks, that is good to know.

If you are suffering a lot of app crashes, the most likely reason is the tablet you are using isn't up to the job.

Labroides Posted at 2017-9-21 23:25
If you are suffering a lot of app crashes, the most likely reason is the tablet you are using isn't up to the job.

And that is a big problem when the tablet is a integrated part of the controller on the minus (+) models! :-(

wataero Posted at 2017-9-21 13:19
That is correct only if you are not in an intelligent flight mode such as POI, where the joysticks are reprogrammed.  When my P4P+ is in POI and circling and the GO app crashes it continues in POI mode.  You can pause it, with the pause button, but there is no published method that I have seen for getting it out of POI so it can be flown home with the joysticks.   It has been suggested to briefly put it in "S"  mode but there is nothing published about whether that would erase the home point and RTH info.  Can you verify that it does not?  Also, is there some other trick to reboot the display such as holding display button down for 10 seconds or something?   Otherwise the only way to get the GO app back is to turn RC off and on.

When the app on mine has f*cked up i can usually just tap the upper left corner and exit to the main screen from where i can press "Go Fly" again to start app once more.

Edit: this also would avoid RTH as the controller keep in contact with AC.

KEDDK Posted at 2017-9-22 03:35
When the app on mine has f*cked up i can usually just tap the upper left corner and exit to the main screen from where i can press "Go Fly" again to start app once more.

Edit: this also would avoid RTH as the controller keep in contact with AC.

Unfortunately I have the PRO+ with the embedded display and when it crashes it goes completely dark, no Android OS screen and no way to reboot display without turning RC off and on.  

wataero Posted at 2017-9-23 18:19
Unfortunately I have the PRO+ with the embedded display and when it crashes it goes completely dark, no Android OS screen and no way to reboot display without turning RC off and on.

I have the same one, but i have not seen it go black like that at all.
It sometimes would tell me that Go4 has stopped responding and i have the choice to stop it. I have had some times the video feed has stopped/gone frozen or other things go cray and i can close it myself and start it again from "Go Fly".
I have not yet had to turn it completely off, but as things are evolving i guess it is just a matter of time ... :-/

we had similar problems including a complete loss of control with a loss of drone, we made the necessary in our country with a occurrence report to the OFAC as well as to the two main helicopter companies, during demonstration it was current to have a drone that made images, this is now completely forbidden

Olivier Vietti- Posted at 2017-9-23 20:50
we had similar problems including a complete loss of control with a loss of drone, we made the necessary in our country with a occurrence report to the OFAC as well as to the two main helicopter companies, during demonstration it was current to have a drone that made images, this is now completely forbidden

We know ... you have posts all over the forum telling that you lost a drone.
But you aren't willing to post any evidence so your claim is very dubious.

but it is not a hazzard, although it only affects a few people, the bug exists, I have to say else found a bug with the clock, it is therefore possible that the syncronization gps does not just right and misleads the drone, attached the file of the flight in question impossible to link the flight "blocked by dji"

https://www.dropbox.com/s/qcjdz8 ... 9-35-54% 5D.txt? dl = 0

Olivier Vietti- Posted at 2017-9-23 23:17
but it is not a hazzard, although it only affects a few people, the bug exists, I have to say else found a bug with the clock, it is therefore possible that the syncronization gps does not just right and misleads the drone, attached the file of the flight in question impossible to link the flight "blocked by dji"

https://www.dropbox.com/s/qcjdz8 ... 9-35-54% 5D.txt? dl = 0

That makes interesting reading.
I'll try to explain what I see in the flight data that DJI didn't explain very well.
To go back to your original report of the loss of the P4 pro, you said...
Lastly during an exercise with the fire brigade, the connection with the drone was lost without possibility of recalling the drone, which left full south at high speed. The incident was submitted to DJI, who today writes that they see a loss of connection with the indication of the 77% battery and the position of the sleeves. DJI offers a discount of 30% in the form of coupon, it is unacceptable, I think that any machine can have unfortunately a computer problem or hidden hardware that can not be detected by the end user. I am reassured that this drone apparently did not cause damage to third parties on the ground ... We are also strongly disappointed DJI after-sales service. I hope this will not happen to you in your life.

Looking at your flight record, it shows that you flew out over a large lake and had it flying in P-GPS and Sport modes.
The last few seconds of the flight record are what is important because up till then everything is quite normal and there are no error indications.

At 7:50.5 the Phantom is 3600 feet out on the lake and flying at 35 mp in Sport Mode.
The IMU altitude is showing 18.7 feet altitude above launch point which was at or close to water level.
you have been flying at moderate fast speed and push the right stick hard forward and quickly get to 40 mph.
At 7:54 when the the flight record ends you still have the right stick hard forward.

Looking back at your altitude readings, there are some big differences between the IMU altitude and the VPS altitude.
At 7:50.5 you had IMU alt = 18.7 ft    VPS = 11.5 ft
At 7:53.6         IMU = 20.7 ft   VPS = 6.2 ft

It's usual for the IMU altitude accuracy to vary as the Phantom warms up during a flight so we can not rely on the IMU altitude to give an accurate height for the Phantom.
We also know that VPS is not accurate for use over water or reflective surfaces and is strongly affected by the tilt of the Phantom when flying fast.
Your Phantom was flying with a pitch angle of 35°-38° just before the flight record ends.
The specs tell us that the VPS has a velocity range of ≤31 mph (50 kph) at 6.6 ft (2 m) above ground.
This is because at greater speeds, the tilt of the Phantom makes the VPS altitude measurement inaccurate as it is designed to give a reading of the vertical distance below the Phantom.

Page 29 of the P4 pro manual gives a list of conditions where pilots should use caution with regard to the VPS:
Operate  the  aircraft  with great caution in the following situations:
a) Flying over monochrome surfaces (e.g. pure black, pure white, pure red, pure green).
b) Flying over a highly reflective surfaces.
c) Flying at high speeds of over 31mph (50kph) at 2 meters or over 11mph (18kph) at 1 meter.
d) Flying over water or transparent surfaces.
e) Flying over moving surfaces or objects.
f) Flying in an area where the lighting changes frequently or drastically.
g) Flying over extremely dark (lux < 10) or bright (lux > 100,000) surfaces.
h) Flying over surfaces that can absorb sound waves (e.g. thick carpet).
i) Flying over surfaces without clear patterns or texture.
j) Flying over surfaces with identical repeating patterns or textures (e.g. tiling).
k) Flying over inclined surfaces that will deflect sound waves away from the aircraft

In another post you said ....
In fact, there were just 30 firefighters who saw the drone sailing at high speed due south while I had my hands placed towards the sky to show that I no longer had control of this drone, and for the data they are in the remote control and the recording stops where the drone takes flight ...

Perhaps Swiss firefighters all have exceptional eyesight but I'd still be surprised if they can clearly see a Phantom close to the lake's surface at a range of 1100 metres.
I mostly fly over water and I know that I have trouble keeping a clear view at half that distance.
Clearly some of the incident description does not match the story that the data is telling us.

The data is more reliable than the opinion of someone that is unsure of what they are seeing and/or doing.

So what we can be sure of is:
You flying very fast, very close to to the lake's surface, 1100 metres out.
The altitude indications cannot be relied on for accuracy but they do tell us that you were very close to the water.
The Phantom appears to be responding to your control inputs.
See 7:16 when you changed direction and brought the drone lower for an example.

We regret that Dji makes no guarantee, we have decided to inform our partners of the low reliability of your products in order to prevent an accident and especially to avoid our bad experience Dji ...
Perhaps you should inform your partners that they should not allow you to fly drones for them.
Sometimes the flight record shows evidence of a malfunction.  
Sometimes it doesn't.  This is one of those times.
It appears that the problem was the operator and not the drone.
It is most likely that you "lost control" of the drone because you flew it into the lake at high speed and too far out to be able to see it.

Labroides Posted at 2017-9-24 00:48
That makes interesting reading.
I'll try to explain what I see in the flight data that DJI didn't explain very well.
To go back to your original report of the loss of the P4 pro, you said...

as you can see there are two intervention boats, we had to follow one while the other was following the drone, and as you will see it was quite high, but I think you will find another good explanation to put the need to the pilot who to more than 10 years of modeling helicopter to his credit

https://www.skypixel.com/videos/ ... ;utm_campaign=share

Olivier Vietti- Posted at 2017-9-24 06:45
as you can see there are two intervention boats, we had to follow one while the other was following the drone, and as you will see it was quite high, but I think you will find another good explanation to put the need to the pilot who to more than 10 years of modeling helicopter to his credit

https://www.skypixel.com/videos/look-at-this-amazing-piece-i-shot-with-my-phantom-4-pro-8455730a-ab35-45c0-a4ef-0bb672a691e2?utm_source=url&utm_medium=copied&utm_campaign=share

Here's what your flight data looks like for anyone interested:
http://www.phantomhelp.com/LogViewer/2ZR9YSZUXAKYXHJLI78D/

As you will see it was quite high, but I think you will find another good explanation

Quite high??   Really??
I can't see that it was quite high at all.

For the last 35 seconds you had brought the Phantom down close to the water.
The IMU barometer (which is only approximate) was showing less than 20 feet and you were low enought for the VPS (also approximate) to indicate a height of 6ft.
You were flying in Sport Mode at 40 mph and out of sight 1100 metres away and within a few metres of the water.
What could possibly go wrong?

As I see it the data tells a very different story from what you have.
I've double checked and I can't find anything that I would change in what I wrote.
I believe that the only explanation that fits the data is that you flew your Phantom into the lake at high speed.
I can't see any other good explanation
If you are sure there is an error in my data analysis, please explain what I got wrong?

If you have the Plus remote control, and you turn OFF the remote control. After 10 seconds of signal loss, the drone will begin to initiate RTH. Then turn your remote back on and try to reinstate drone communication through the go 4 app. I first purchased the Phantom 4 Pro and later bought a Plus remote control. I used it for about 2 1/2 weeks then returned it to BH Photo for 100% refund. I returned it under my experience that it "Did Not Perform as Expected". When possible I make my expensive purchases from places with a satisfaction 30 day return policy. With that being said, Fry's in Renton Washington has a Zero return policy on Drones. If it has a factory defect you can exchange for the same model, but that's it. All models from toy's to professional models. Keep that in mind when buying a drone. You may want to buy from BH Photo or Adorama, in the USA.

Labroides Posted at 2017-9-24 11:46
Here's what your flight data looks like for anyone interested:
http://www.phantomhelp.com/LogViewer/2ZR9YSZUXAKYXHJLI78D/

if you were able to download the cached video you will see that visuellement at the time the connection was completely lost we made to ascend the drone, then as we had more return of the drone and that the application of the drone had crashed and that the led was red we pressed on the button rht but nothing we switched off then on again and still no connection and we remade the rth button and in connection with the crews on the water the intervention leader who was with me and the crew of reserve on the ground had the return of those who I filamt that the drone was sailing full south at high speed, we organize a beat on the shores of the lake where our assemblies begin to find it but without success
https://www.skypixel.com/videos/ ... ;utm_campaign=share

MrDave Posted at 2017-9-24 16:46
If you have the Plus remote control, and you turn OFF the remote control. After 10 seconds of signal loss, the drone will begin to initiate RTH. ...

After TEN seconds?  I didn't know that.  Good to know - thanks

Mark The Droner Posted at 2017-9-25 02:24
After TEN seconds?  I didn't know that.  Good to know - thanks

It's not 10 seconds at all.
It's three seconds  just as it says in the manual.
This forum is very good for spreading false information.

MrDave Posted at 2017-9-24 16:46
If you have the Plus remote control, and you turn OFF the remote control. After 10 seconds of signal loss, the drone will begin to initiate RTH. Then turn your remote back on and try to reinstate drone communication through the go 4 app. I first purchased the Phantom 4 Pro and later bought a Plus remote control. I used it for about 2 1/2 weeks then returned it to BH Photo for 100% refund. I returned it under my experience that it "Did Not Perform as Expected". When possible I make my expensive purchases from places with a satisfaction 30 day return policy. With that being said, Fry's in Renton Washington has a Zero return policy on Drones. If it has a factory defect you can exchange for the same model, but that's it. All models from toy's to professional models. Keep that in mind when buying a drone. You may want to buy from BH Photo or Adorama, in the USA.

It’s 10 seconds when using spark or Mavic with phone only, but when using RC 3 seconds , and you have to use RC with P4Pro so when you lose signal with P4Pro as pointed out RTH will be initiated after 3 seconds.

Thank you both.  

wataero Posted at 2017-9-21 13:19
That is correct only if you are not in an intelligent flight mode such as POI, where the joysticks are reprogrammed.  When my P4P+ is in POI and circling and the GO app crashes it continues in POI mode.  You can pause it, with the pause button, but there is no published method that I have seen for getting it out of POI so it can be flown home with the joysticks.   It has been suggested to briefly put it in "S"  mode but there is nothing published about whether that would erase the home point and RTH info.  Can you verify that it does not?  Also, is there some other trick to reboot the display such as holding display button down for 10 seconds or something?   Otherwise the only way to get the GO app back is to turn RC off and on.

The intelligent flight mode only works in P mode. You can switch the flight mode and back to P mode, the drone will exit the POI.

Olivier Vietti- Posted at 2017-9-24 19:02
if you were able to download the cached video you will see that visuellement at the time the connection was completely lost we made to ascend the drone, then as we had more return of the drone and that the application of the drone had crashed and that the led was red we pressed on the button rht but nothing we switched off then on again and still no connection and we remade the rth button and in connection with the crews on the water the intervention leader who was with me and the crew of reserve on the ground had the return of those who I filamt that the drone was sailing full south at high speed, we organize a beat on the shores of the lake where our assemblies begin to find it but without success
https://www.skypixel.com/videos/look-at-this-amazing-piece-i-shot-with-my-phantom-4-pro-8455730a-ab35-45c0-a4ef-0bb672a691e2?utm_source=url&utm_medium=copied&utm_campaign=share

I've independently taken a very close look at this flight and video.  The flight record shows no discrepancies aside from altitude at the very end.  I completely concur with what Labroides has written in post #21.

The contentious piece starts from 7 minutes 41 seconds onwards.  At this point the VPS and barometric altitudes are more or less in sync at 15-17 feet.  The home point is only 1 meter above the water level (372m ASL) and that may contribute to the small discrepancy.  From the video, I estimated the camera's horizontal plane to be the height of three passengers  standing in the boat stacked on top of each other (approx. 6m), which corresponds nicely with the altitude records.

The pilot then began chasing a RIB, an inflatable fast response vessel complete with aerofoil on the rear.  This raised bar and warning beacon lights, although not a wing, will still act as a spoiler and disrupt air, producing vortices and bad air behind the boat.  This is in addition to the disruption caused by the bow of the boat lifting into the air at high speed.

The pilot flew on a course of about 145°, while facing the aircraft approximately East to view the boat.  He took up a position relative to the RIB at approximately 45° behind and starboard of the vessel.  Both vehicles accelerate up to 18m/s (40mph) at the end of the record.  At low speed the flight is unremarkable.  At high speed you can see the nose of the boat lift and begin to see the effect from an air vortex from the bow of the craft all along the side, appearing and disappearing  quickly, adding to the more constant water wake.  This wake turbulence appears to spread out at a 45° angle from the bow.  The chase continues for a further 13 seconds until the flight record runs out.  The video however continues momentarily longer and we see the aircraft ascend slightly before the video pixelates and freezes, indicating a loss of signal.  This rise was not commanded by you and there is nothing in the log to show anything other than full forward acceleration in Sport mode.  I postulate that the aircraft was in an unfortunate position to be following a high speed vehicle and was caught in the tip of the air vortex coming from the craft.  It would have been safer to have recorded the sequence from a greater altitude or further forward or further to the side.  An aircraft pilot would never follow behind another aircraft at that angle in that zone as the vortex would disrupt his lift and flight.  This appears to be exactly what we see here.

Example pictures, showing the effect from aircraft which is just as valid for a boat.

I believe we see the aircraft ascend slightly before the loss of video as it is being disrupted by a vortex and lifted in this "death spiral".  It is likely the camera gimbal did its job and does not reflect the tumbling of the aircraft at this stage.  This altitude rise is not reflected in the flight log, so the data was not successfully transmitted at this point in time, but the video was (opposite to what we usually see).   What we do see in the flight record is the VPS altitude reading go from 6 feet to 0 instantly.  This could well be explained by the aircraft rolling to its side, no longer allowing the ultrasonic transducer to get a reading off the surface of the water.  Indeed what we see its the aircraft rolling from -3° to +4° (and presumably beyond when the telemetry is lost).  This would result in the aircraft performing a clockwise roll (top to starboard).  In this position the aircraft body would be obstructing the legs (antennas) from a line to the controller on the shore.  This would cut telemetry and video transmission, and in this configuration would no longer be providing lift.  At such a low altitude it is probable that the  aircraft was spun by the vortex straight into the water and was lost immediately.  This explains why there is no resumption of telemetry or restoration of video.  It would all have happened very quickly.

Choosing to fly in that position behind a fast moving craft is probably pilot error number one.  Throwing your hands in the air when you lose control and not fighting as hard as you can to recover your aircraft would be error number two.  Hijacking multiple posts on the forum lamenting the loss of your aircraft and alleging hardware and software errors for your misfortune is just bad form and disrespectful.

Sincerely, I am sorry for your loss, but this incident can be explained as pilot error, by flying into a vortex produced by a larger and much more powerful vessel.  This can be confirmed by releasing smoke from the rear of the RIB at high speed, and observing the shapes produced.

endotherm Posted at 2017-9-27 12:09
I've independently taken a very close look at this flight and video.  The flight record shows no discrepancies aside from altitude at the very end.  I completely concur with what Labroides has written in post #21.

The contentious piece starts from 7 minutes 41 seconds onwards.  At this point the VPS and barometric altitudes are more or less in sync at 15-17 feet.  The home point is only 1 meter above the water level (372m ASL) and that may contribute to the small discrepancy.  From the video, I estimated the camera's horizontal plane to be the height of three passengers  standing in the boat stacked on top of each other (approx. 6m), which corresponds nicely with the altitude records.

While I think you had a really good stab at what happened, there is something that doesn’t add up, this RIB is no bigger than 5 metre and looking at the engine I would say it’s 75 hp with 4 adults and 2 children on board it’s almost impossible for it to reach 40mph.
I say this as someone who pilots a 6.5 metre rib with a 225 hp engine which has a max speed of 49 mph with one person on board, this speed is much reduced with 4 people on board. So your theory may need some recalculating.

hallmark007 Posted at 2017-9-27 12:36
While I think you had a really good stab at what happened, there is something that doesn’t add up, this RIB is no bigger than 5 metre and looking at the engine I would say it’s 75 hp with 4 adults and 2 children on board it’s almost impossible for it to reach 40mph.
I say this as someone who pilots a 6.5 metre rib with a 225 hp engine which has a max speed of 49 mph with one person on board, this speed is much reduced with 4 people on board. So your theory may need some recalculating.

Without getting into the performance of the particular vessel, selected for rescue duties where speed and load carrying capabilities is paramount, I have merely extrapolated the data.  The aircraft was keeping a constant distance, and only hit 40mph in the last two seconds of acceleration.  The last few frames show it washing off speed as the pitch changes from -30° (full nose down) and the prop disappears from view.

In this case, watching the video puts the data in context.  Analysing the data alone would not have revealed any fault at all, and this would have been one of those "head-scratchers".  Seeing the video confirms the presence of an external influence that is not reflected in the logs.

endotherm Posted at 2017-9-27 12:42
Without getting into the performance of the particular vessel, selected for rescue duties where speed and load carrying capabilities is paramount, I have merely extrapolated the data.  The aircraft was keeping a constant distance, and only hit 40mph in the last two seconds of acceleration.  The last few frames show it washing off speed as the pitch changes from -30° (full nose down) and the prop disappears from view.

In this case, watching the video puts the data in context.  Analysing the data alone would not have revealed any fault at all, and this would have been one of those "head-scratchers".  Seeing the video confirms the presence of an external influence that is not reflected in the logs.

I think we have wasted enough time, it is understood that you will always find good escuses to not take over the guarantee, it is a pity that the follower boat that monitored the drone as agreed and that was also to join the to begin the second exercise to see your drone from full south then it is the lead boat that was surpassed when this one to slow down, we are in the helicopters as a professional and we know perfectly the downwash, and I confirm that after losing the video signal to push the pitch to make the drone up to find the video return, but nothing, we acted the rth of the command, but there nothing either, is on is that the occurrence report completed by the intervention leader is now known by the OFAC as well as by our company as well as our Vaud colleagues.
it is a shame, you have super products, a pity that your after sales service is unfortunately pitiful, to see the number of cases reported on your forum our experiences and those of others will pay attention to potential buyers for a dji or another manufacturer. ..

hallmark007 Posted at 2017-9-27 12:36
While I think you had a really good stab at what happened, there is something that doesn’t add up, this RIB is no bigger than 5 metre and looking at the engine I would say it’s 75 hp with 4 adults and 2 children on board it’s almost impossible for it to reach 40mph.
I say this as someone who pilots a 6.5 metre rib with a 225 hp engine which has a max speed of 49 mph with one person on board, this speed is much reduced with 4 people on board. So your theory may need some recalculating.

I am being pedantic, but isn't it actually a RHIB, not RIB. As in Rigid Hull Inflatable Boat.
OK, I was on my way out the door.....

Geebax Posted at 2017-9-27 18:13
I am being pedantic, but isn't it actually a RHIB, not RIB. As in Rigid Hull Inflatable Boat.
OK, I was on my way out the door.....

Mine would be hyphenated, a Rigid-hulled Inflatable Boat, so the acronym still stands .  I don't mind pedants... though you are correct.

Olivier Vietti- Posted at 2017-9-27 17:51
I think we have wasted enough time, it is understood that you will always find good escuses to not take over the guarantee, it is a pity that the follower boat that monitored the drone as agreed and that was also to join the to begin the second exercise to see your drone from full south then it is the lead boat that was surpassed when this one to slow down, we are in the helicopters as a professional and we know perfectly the downwash, and I confirm that after losing the video signal to push the pitch to make the drone up to find the video return, but nothing, we acted the rth of the command, but there nothing either, is on is that the occurrence report completed by the intervention leader is now known by the OFAC as well as by our company as well as our Vaud colleagues.
it is a shame, you have super products, a pity that your after sales service is unfortunately pitiful, to see the number of cases reported on your forum our experiences and those of others will pay attention to potential buyers for a dji or another manufacturer. ..

This is a public forum.  The majority of responses are from fellow pilots, with nothing to gain from their opinions and input.  My analysis is completely independent from any one else including DJI.  It is based on experience and observation, including a career of accident investigation and data analysis.  I do not work for DJI.  I am not making excuses for DJI not to honor their warranty, but you haven't identified any fault at all.  The logs simply do not support any claim of a fault.  Flying a 1.3kg machine so close to a powerful vessel is asking for trouble, and the video shows that you did indeed fly in a hazardous position.  Trying to climb or press RTH will be useless if the aircraft has already been dumped into the water.  Reporting the incident to your civil aviation authority (The Federal Office of Civil Aviation [FOCA]) is not evidence of a fault or defect in the aircraft.  Many experienced commercial pilots have been caught out by unseen factors such as wind-shear that they may not have experienced before, some resulting in loss of aircraft.  It happens to the best of pilots.  We all learn from this.  Next time, we will all be aware not to fly so low and close to a fast moving vehicle, thanks to your loss.

Geebax Posted at 2017-9-27 18:13
I am being pedantic, but isn't it actually a RHIB, not RIB. As in Rigid Hull Inflatable Boat.
OK, I was on my way out the door.....

Yes you are correct,it is RHIB, but it is much more commonly know by RIB, guess I thought there was a lot more common people around here. It’s hard to say RIB when you put a H in the middle of it, maybe with the the eloquent Aussie accent it works. Lol

Geebax Posted at 2017-9-27 18:13
I am being pedantic, but isn't it actually a RHIB, not RIB. As in Rigid Hull Inflatable Boat.
OK, I was on my way out the door.....

So Stewie, "Why are you putting so much emphasis on the 'H'?"
(From Family Guy).

DJI Susan Posted at 2017-9-27 00:19
The intelligent flight mode only works in P mode. You can switch the flight mode and back to P mode, the drone will exit the POI.

I have been reading responses to my problem of GO app crashing in intelligent flight mode.  While I would like to see DJI get rid of the app crashes it is unlikely to never happen again.  So I need a procedure to follow and based on comments I came up with the following:
1) Hit pause button and observe aircraft to see if it did pause and go into hover.
2) Switch to ATTI mode and back to “P” mode to get it out of intelligent flight mode.
3) If aircraft is close by manually fly it home.
4) If aircraft farther away and if I’m sure that smart RTH has been set up properly and area is clear, then press button to start RTH.
5) If aircraft starts to ascend to RTH altitude, then take hands off controls unless needed for landing.
6) If nothing happens when RTH button is pushed then turn RC off & on and when RC link light is green start  GO app.  When GO app shows aircraft status and location, try RTH again and if it fails then fly home manually.
Is there anything I’m missing, suggestions?