Just throwing my hat in the ring of flyaways and crashes.  I had half a dozen or so great flights with my first Inspire 1 (I own two), got some great video and never had a hiccup in performance.  Took it to Joshua Tree, CA for some scenery shoots.  It is an unpopulated area with no man-made structures or powerlines except for a 2-lane road, nobody was around.  Placed aircraft on a flat, level, paved surface, powered up, exited travel mode, and powered down.  Attached propellers and camera, activated remote and connected iPhone, activated DJI Pilot app, powered up the aircraft.  After gimbal and compass calibration, status light showed green, and remote showed good GPS connection with 17 satellites.  Aircraft, remote, and iPhone batteries were all at 100% at time of takeoff.  Weather was approximately 75F with a steady wind speed of approximately 5mph with no gusting, clear visibility.
Lifted off the ground at approximately 5:11pm and headed toward a nearby rock pile to shoot video at a range of approximately 650-1100ft from home point, maintaining visual contact with the aircraft.  Three successful video shoots included several 360degree rotations around the rock pile without losing remote connection or video integrity.  Full bars of connectivity with the remote and the video feed were noted throughout the flight, as well as consistent connection with approximately 17 GPS satellites at all times, and the aircraft was never obscured from the sky.  The fourth video shoot commenced at approximately 7 min: 20 sec after takeoff.  While flying forward toward the rock pile, the remote lost connection without warning and did not regain connection with the aircraft, although the remote remained powered on.  All telemetry data froze as though the aircraft had been powered down or disconnected.  At the time communication was lost, the aircraft was headed toward the rock pile at a rate of 11.4mph at a height of 105.0ft and a distance of 969.8ft from home base, and was visible to the pilot. At this time, the home point indicator was green on the DJI Pilot app for the location of the remote control (where I was standing) and the aircraft battery was at 49%.
The aircraft continued behind the rock pile out of sight and did not ascend and return home.  At this time, I ran toward the rock pile with the remote attempting to locate the aircraft and reestablish remote connection.  While running I had the left stick at full upward, so that if connection was reestablished, the aircraft would ascend away from any obstacles.
The aircraft was found upside down at the base of the rock pile with the landing legs and propellers destroyed, the camera and gimbal separated, and the aircraft body moderately damaged.  It was apparent that the aircraft had flown laterally into the rock pile at a high rate of speed and fallen from an unknown but significant height onto the rocks below.  The remote control battery and iPhone battery were both above 90% throughout the flight, and the aircraft battery was powered on when the aircraft was recovered, with an indicated charge of 49%.
I sent the aircraft to DJI for repair, and will likely have to wait a month or two to hear anything or get it back, which sucks - but they are apparently understaffed for the flood of repair orders coming in.  I feel I am owed a replacement due to the failure of the Failsafe RTH function when the remote lost connectivity, not to mention that the remote lost connectivity in a broad open space with a clear line of sight and no metallic or electrical structures nearby.  
As a note to DJI, I really wish they put together a better repair/replacement turnaround for the hoards of owners of $3000 aircraft that fail on basic software or hardware glitches.  My other Inspire 1 suffered a camera failure five minutes out of the box after two successful video shoots during the first flight, where all flight and telemetry feedback are normal but the camera feed is black with a "No Signal" status.  That one also has and RMA with an estimated turnaround of 6 weeks, just to replace the camera with an obvious manufacturing defect (all troubleshooting efforts yielded no improvement).  If DJI is going to beta test their products on their customers, they should be prepared to fix the bugs and issues.

Any thoughts?

Hi
Do You have a video? What abot the sensors? Have You checked them?
That's realy sad to read such stories.

Sorry to say but it sounds like you lost video feed not remote control feed. If you still had a green light on your remote then it was still talking to the inspire. That is way a RTH was never activated. You may have drove the bird into something when you thought you lost connection.

  

lkedziora@gmail Posted at 2015-6-5 04:08
Hi
Do You have a video? What abot the sensors? Have You checked them?
That's realy sad to read such ...

The video I was in the process of recording at the time of the crash was not completed and saved properly, so I can't open it.  But either way it's in the hands of DJI now.

tapboxer@gmail. Posted at 2015-6-5 04:13
Sorry to say but it sounds like you lost video feed not remote control feed. If you still had a gree ...

I don't think so.  Because I could still see the aircraft when I lost connection, and tried to stop it and up-stick to climb to a position I could get control back, and it continued to fly behind the rocks.  Also in my in-flight data record it abruptly cuts off while the aircraft is still in the open area before it disappears behind the rocks, and at the point it cuts out the controller is on a gentle throttle-up and not veering toward any obstacles.  Also, the video feed wasn't the only thing that went away; everything froze including the telemetry data, as though I had hit the power switch on the aircraft.  When I say the home point was green, I meant that was at the time I lost connection - once I lost connection everything went MIA.

I am sorry for your loss Bro! I sent my Inspire to torrance Ca. for repair on june 3rd they received it may be back in my hands by september.

CrabHawk Posted at 2015-6-5 05:22
I am sorry for your loss Bro! I sent my Inspire to torrance Ca. for repair on june 3rd they received ...

Yeah it sucks when you have a $3000 investment that works for an hour and then you're waiting three months for another chance to use it.

range of approximately 650-1100ft from home point, maintaining visual contact with the aircraft

I apologise in advance if I sound like an arse. Can you really see orientation with LOS from that distance? Even at 200m I find it hard to see the orientation of the aircraft as it is becoming a dot in the sky. Return To Home is a fail safe. A last resort if you will. In the UK the CAA suggest a maximum of 500m or less if you lose visual orientation (i.e. looking at the aircraft, NOT at the video screen).

There is a aspect here, an elephant in the room as it were here. These are highly complex machines that are subject to all sorts of interferences and influences that the user may not be aware of. We know they are not 100% reliable. There is no single flight controller on any UAV, be it a Pix Hawk, Ardu Pilot, or DJI, that is 100% reliable in all circumstances. They have all suffered flyaways in some form.

Therefore, unfortunately, people should realise that they should be flying these machines responsibly. Just because they can go to a certain range, that does not mean that you should. There is a fundemental thought that should run through your head whenever you fly. And that thought is "what will I do if I have a problem?" You should assume that you will have a problem and have a plan in place. You should be happy that a crash will be safe for those around. But you should also be ready with a procedure to run through.

In this instance did you switch immediately to ATTI mode at the first sign of trouble? Communication is through the transmitter, not the app. If there was a conflict between sensors on the aircraft, you should switch immediately to ATTI mode and attempt to regain control. It should be your first reaction. If this doesn't work you should attempt to cut all power. If this doesn't wrk and your aircraft really does just fly off on its own then you should be aware of the airspace you are flying in and be ready to alert relevant authorities. The Inspire isn't a toy. It shouldn't be treated as such. It is a device that could cause serious injury and should always be flown as such.

Sorry if that sounds overly serious and stern, but I am afraid that many of these flyaways do not state if any of the above actions were taken so we cannot be sure that the flyaway or crash was due to a real malfunction, or because the pilot was inexperienced and left the craft in GPS mode while it was all going wrong.

CrabHawk Posted at 2015-6-5 05:22
I am sorry for your loss Bro! I sent my Inspire to torrance Ca. for repair on june 3rd they received ...

they received mine on May 18th. they told me 6 weeks this is week 3. they said it would be good as new or better when I get it back. my problem was rc controller would not except firmware upgrade and I had no video feed to my tablet any more. "no crash"

SimonMW Posted at 2015-6-5 06:10
I apologise in advance if I sound like an arse. Can you really see orientation with LOS from that d ...

Once you've lost the remote link, I don't see how the Atti Mode initiation command will be sent to the A/C or how it would work if the A/C was somehow triggered into Atti because you don't have control.

SimonMW Posted at 2015-6-5 06:10
I apologise in advance if I sound like an arse. Can you really see orientation with LOS from that d ...

I agree with what you're saying, but most of it doesn't apply to this situation (aside from the general safety stuff, which always applies, and I went to the trouble in my original post of detailing all the safety precautions I took such as flying in an area with no people or buildings, calibrating, etc).  Telling me the aircraft is "not a toy" and should be flown "responsibly" suggests you didn't read the original post thoroughly, and are responding based on assumptions.

The fact is, the remote control lost its connection.  In addition to losing video feed, no stick inputs were accepted.  I know this because when I lost video feed, I could still see the aircraft, and I stopped forward impulse and gave it full upward throttle, but it continued flying straight forward toward the rocks without accepting my input.  As for the aircraft's orientation, I knew the direction the craft was heading at the time I lost control, and when I stopped forward impulse and gave it full upward throttle, the orientation at that point did not matter, as the aircraft was clear of all obstacles and had nothing above it.  If it had accepted my inputs, it would have flown upward (regardless of whether it was in ATTI mode and/or drifting laterally).  

Yes, of course the Failsafe RTH is a last resort.  Anyone with even a basic knowledge of flight safety understands that autopilot functions are not to be relied upon as a primary means of operating the aircraft.  But in the event that the remote connection is completely dropped, that is exactly why the Failsafe RTH was designed, and when it does not operate and the craft crashes due to loss of remote control signal, then the last resort functions do not do any good.

Just like any other safety feature - the airbags in a car will hopefully never be used and should not be relied upon in place of safe driving and situation awareness, but if the situation arises in which they are needed, they should operate as designed.  Please read the original post for a detailed listing of the safety precautions taken before and during the flight that would have been sufficient to avoid a crash if the equipment were operating as designed.

Hi
Would You please send the link to Google Map of the spot You were flying?
Thank You in advance

Once you've lost the remote link, I don't see how the Atti Mode initiation command will be sent to the A/C

Switching to ATTI mode using the switch on the transmitter would have been a first step in quickly diagnosing what was going on. Can't make assumptions. It should have been the first reaction. If it still didn't work then other steps have to be taken. Going to ATTI would show whether the aircraft was going haywire on its own due to sensor conflicts, or whether there was a true loss of control. I have had blackouts on the iPad app, and it has said it has lost connection to the aircraft, but the aircraft has always remained under the control of the physical TX. So was it just the app saying it had lost connection, or had the TX really lost connection with the aircraft (what colour was the power light on the TX?)? You would never know unless you hit that ATTI switch.

What were the sensor mod values before take off? Did you calibrate near to your iPhone or your TX (you said you were using it to run the app, and it is known that phones interfere with calibrations and is mentioned in the manual, even if it did tell you the calibration went fine.) Did the compass reading match the physical orientation of the aircraft before take off? Did the aircraft think it was automatically initiating RTH with some sort of sensor conflict that sent it in the wrong direction? Did you calibrate your IMU after firmware upgrades?

There are too may factors unanswered to be able to say that this is due to DJI manufacturing.

The fact is, the remote control lost its connection.  In addition to losing video feed, no stick inputs were accepted.

Precisely. You can't fully tell unless you went to ATTI where the sensory control of the aircraft is turned off.

I know I sound like a bit of an arsehole, and for that I apologise. I do not intend to be derogatory or patronising. I would like to know what caused this as well, but every minute stage and process has to be eliminated methodically with no stone unturned. You should see if you can download the flight data to see what went on when it happened, too. This is why many flyaways are impossible to diagnose. There are so many little things that could contribute.

lkedziora@gmail Posted at 2015-6-5 17:56
Hi
Would You please send the link to Google Map of the spot You were flying?
Thank You in advance

https://www.google.com/maps/@34.0891417,-116.2585797,847m/data=!3m1!1e3