Drone did not RTH!

[BudWalker]

There have been fly away incidents with the P3 where Yaw was compromised (i.e. compass error) and RTH was initiated. Probably the same possibility exists with the Mavic. If the AC is launched from a geomagnetically distorted site Yaw can be compromised at launch and remain compromised. If the pilot initiates an RTH in response to the resulting erratic flight and/or compass errors a fly away is almost guaranteed. I looked at one incident where there was an RC disconnect which caused RTH to be initiated and resulted in a fly away. It doesn't seem to matter if the AC is in ATTI mode, RTH can be initiated and the AC will use the compromised YAW to attempt the RTH.

@scjerry please provide a .DAT for one of these incidents and we can take a look.

A compass calibration performed in geomagnetically distorted location that results in a flawed calibration is very unlikely, if not impossible. It's mathematically impossible for a calibration to detect and then compensate for magnetic anomalies external to the AC. Attempting to calibrate in a geomagnetically distorted location will cause the calibration to be rejected. There has been some discussion of this over on PhantomPilots starting here
Compass Calibration, A Complete Primer

 

[NestleQuik]

I'm thinking, and i'm not positive, but maybe you should have calibrated your compass at the new location before taking off. Not sure if that would have helped, but i read its recommended to help prevent fly aways

 

[JeepDSL]

i read its recommended to help prevent fly aways

Here we go again...read page 52 in the Mavic Pro User Manual.

 

[NestleQuik]

Here we go again...read page 52 in the Mavic Pro User Manual.

You don't think calibrating your compass in a new location BEFORE flight is a good idea to potentially prevent a flyaway?
I opened my book again after 10 months of owning my Mavic and Page 52 wasn't relevant to my comment..

 

[JeepDSL]

You don't think calibrating your compass in a new location BEFORE flight is a good idea to potentially prevent a flyaway?
I opened my book again after 10 months of owning my Mavic and Page 52 wasn't relevant to my comment..

Page 52 of User Manual v1.6 (Mavic – Specs, FAQ, Tutorials and Downloads): "Only calibrate the compass when the DJI GO 4 app or the status indicator prompt you to do so."

There's lots of discussion and differing opinions in this and other threads (for example, How many of you do compass calibration before take off?).

No, I don't calibrate in a new location before flying. If I went half way around the world, I might. If you want to calibrate, go ahead; it's your drone and your choice.

 

[NestleQuik]

Page 52 of User Manual v1.6 (Mavic – Specs, FAQ, Tutorials and Downloads): "Only calibrate the compass when the DJI GO 4 app or the status indicator prompt you to do so."

There's lots of discussion and differing opinions in this and other threads (for example, How many of you do compass calibration before take off?).

No, I don't calibrate in a new location before flying. If I went half way around the world, I might. If you want to calibrate, go ahead; it's your drone and your choice.

Yea i read that part, but i don't think it would hurt. From others past experiences,who basically had flyaways, they recommended it. i haven't done mines unless i'm prompted but i generally fly from the same location. Boring but enjoyable hahaha

 

[unexpanded]

I fly within 1200mile radius- all the time in different places and sometimes well above the Arctic circle. I've had MP for a month now, have done 2 firmware updates: up to .1000 now. I calibrated the compass for the very first flight only and literally 0 issues so far. Only compass errors I've gotten so far occur when I boot up my MP inside of my car, for obvious reasons.

 

[Ron_Dog]

I always set my phone into airplane mode when I fly. I've never had it fail to return home. I HAVE seen a MP fly crazy because the operator was in a rush to fly and got airborne BEFORE the MP had connected to enough Satalites. I let mine set for a good 2-3 minutes powered up, before lifting off.Apprently having your phone in airplane mode dosen't effect the GPS? Having your wife call you while your flying,will effect your screen though!

 

[scjerry]

Belief has nothing to do with it. If you want to advance the hypothesis that they are connected then you need to suggest a mechanism for that, and/or back it up with some log file data that shows at least a correlation between unexpected RTH and anomalies in the compass data. You have done neither of those - you just keep asserting an undefined connection with the compass, along with other unrelated explanations.

I'm not interested in belief systems as a retired scientist. Data was presented in another posting and the beginning of this one. Log files were presented in the form of Flight Record screen shots from DJI GO 4. In summary, 1) Behavior of drone was different with and without compass calibration. This behavior was repeated more than once and on differing days and unless I manually intervened, I would have lost my drone (see the last screen shot). 2) W/O calibration, the quad went off twice to an unknown HP (not in the log files or on the map) or back to the launch location, after an attempt (documented) to set the HP to the RC. Point in fact, even though the documented screen message displayed that the HP had been updated to the current RC position, it was NOT recorded even though a little voice said check my map (not cached). 3) With calibration, flight data (not included) shows that HP setting to RC location worked and the RTH path led to the new HP location, not the launch position. 3) In conclusion, recognizing that other factors may contribute to this result, somehow the HP data gets corrupted w/o calibration of the compass, and is properly refreshed with calibration. I attempted to also make certain that the invocation of DJI Go 4 was fresh and did not contribute to the HP data corruption Whether this is an engineering issue or a user issue, a resolution is needed.

I will repeat my flights in a new venue completely devoid of electrical or magnetic interference sometime today. I recognize that this may be unique to my drone and use of my iPad Mini 4 wi-fi + cell. Your results may differ. I've included only a few of the screen shots (again) for your perusal. Note the flight mode annotation "Go Home" to the right of the Mavic icon. On the occasion depicted, I did not calibrate the compass nor was I advised to do so by Go 4. It went wayward twice in the same direction after attempting to set the HP to my controller, also twice. Upon calibration, setting the HP to my RC worked and RTH also worked.

BTW, one responder to my other post noted the following: "When you set the home point to the controller position, did you read the notice that appears asking you to check the distance from the home point that is set when you start the drone? If you've not moved then this should be only a couple of feet. If it's more your device hasn't got a gps fix and it's using whatever it has stored."

I'm not sure about this. But it deserves investigation. First of all, Follow Me mode works for either un-calibrated or calibrated, meaning the GPS on my iPad is working, though the precision is unknown. Again, in the calibrated mode, setting the HP to the RC worked.

 

[BudWalker]

@scjerry can you provide the .txt for this flight? The .DAT would be better, but if you're on Firmware v01.03.1000 that will be a problem. An actual log file is much better than a description and several screen shots.

I'd be very surprised if a compass calibration, or lack of, had anything to do with the behavior you're seeing. It's a myth that a calibration determines and compensates for magnetic anomalies external to the AC. A log file will help with this.

 

[flodi]

I always set my phone into airplane mode when I fly. I've never had it fail to return home. I HAVE seen a MP fly crazy because the operator was in a rush to fly and got airborne BEFORE the MP had connected to enough Satalites. I let mine set for a good 2-3 minutes powered up, before lifting off.Apprently having your phone in airplane mode dosen't effect the GPS? Having your wife call you while your flying,will effect your screen though!

Ron, same here...+95% of my flghts are in areas with no wifi and no Cell data (Agriculture usage)...so I use Airplane mode to save phone battery. The one time i got a bit scared with my mavic was the time I took off in a hurry... it started drifting like crazy but I quickly went up as fast as I could to avoid obstacles until it stabilized (Sure it was bad GPS reception or few satelites). It never happened again since I wait to have at list 13 satelites showing on my screen.

And, yes, I tried several times to change my land location to "my location" while on airplane mode and it worked flawless.... example... sometimes I start a dronedeploy mission and set to hover at the end while I am moving around as the mission runs then I close dronedeploy, start Go4 and set RTH to "me" and wait... fearless. It works so far. (my missions usually don't take the drone farther than 1700ft away form me)

As for other settings... VPS and Land protection are off (as i take off and land with my hands because of dust and dirty). I enable VPS just in case I need a more steady hovering to analyze a terrain spot and then disable to RTH

Using an Iphone 7+ (before it was an iphone 6), always the latest dji firwmare and a pinch of ignorance and confidence as I am always flying in open areas with very few risk of collision.

PS. Sorry about the poor english... I am a portuguese speaker

 

[sar104]

I'm not interested in belief systems as a retired scientist. Data was presented in another posting and the beginning of this one. Log files were presented in the form of Flight Record screen shots from DJI GO 4. In summary, 1) Behavior of drone was different with and without compass calibration. This behavior was repeated more than once and on differing days and unless I manually intervened, I would have lost my drone (see the last screen shot). 2) W/O calibration, the quad went off twice to an unknown HP (not in the log files or on the map) or back to the launch location, after an attempt (documented) to set the HP to the RC. Point in fact, even though the documented screen message displayed that the HP had been updated to the current RC position, it was NOT recorded even though a little voice said check my map (not cached). 3) With calibration, flight data (not included) shows that HP setting to RC location worked and the RTH path led to the new HP location, not the launch position. 3) In conclusion, recognizing that other factors may contribute to this result, somehow the HP data gets corrupted w/o calibration of the compass, and is properly refreshed with calibration. I attempted to also make certain that the invocation of DJI Go 4 was fresh and did not contribute to the HP data corruption Whether this is an engineering issue or a user issue, a resolution is needed.

I will repeat my flights in a new venue completely devoid of electrical or magnetic interference sometime today. I recognize that this may be unique to my drone and use of my iPad Mini 4 wi-fi + cell. Your results may differ. I've included only a few of the screen shots (again) for your perusal. Note the flight mode annotation "Go Home" to the right of the Mavic icon. On the occasion depicted, I did not calibrate the compass nor was I advised to do so by Go 4. It went wayward twice in the same direction after attempting to set the HP to my controller, also twice. Upon calibration, setting the HP to my RC worked and RTH also worked.

BTW, one responder to my other post noted the following: "When you set the home point to the controller position, did you read the notice that appears asking you to check the distance from the home point that is set when you start the drone? If you've not moved then this should be only a couple of feet. If it's more your device hasn't got a gps fix and it's using whatever it has stored."

I'm not sure about this. But it deserves investigation. First of all, Follow Me mode works for either un-calibrated or calibrated, meaning the GPS on my iPad is working, though the precision is unknown. Again, in the calibrated mode, setting the HP to the RC worked.

View attachment 20695 View attachment 20696 View attachment 20697 View attachment 20698 View attachment 20699 View attachment 20700

OK - to summarize the problem: When you dynamically set the home point to the RC location, then the aircraft sometimes heads to a different location on RTH. Based on more than one test (how many?) you see a correlation between that behavior and whether or not you just calibrated the compass. Is that correct?

Your hypothesis is that there is a connection between the compass calibration process and the ability of the aircraft to store a home point - correct?

In your screen shots the recorded home point is clearly in the region of 400 m south east of your position, based on the distance readout. That distance should have been what you checked on the reset, and would have been an immediate warning that the the mobile device had supplied a spurious location.

Given that the home point has no compass calibration data, or even bearing data, related to it - it's just a position - why would you not hypothesize first that the cause of the problem is a bad GPS fix on your mobile device, since that is where the dynamic home point position data are coming from? That's a much simpler explanation that you have not ruled out and that does not require some extraordinary mess up in the programming and internal data structures in the FC. My hypothesis would be that the correlation you saw a couple of times with compass calibration is just coincidence.

As pointed out by others, without the flight log or, preferably, the DAT file, it is impossible to be completely definitive, although this is such a relatively common problem (bad initial home point on the aircraft or bad dynamic home point data from the mobile device) that the answer looks pretty clear to me. Take a look over on PhantomPilots to see how many times this has come up in the past.

 

[Noob ie]

Used the "take off" and checked "precision landing" on the DJI Go 4 app. The app confirmed "home point has been updated", the aircraft launched as normal. I was flying in a new area but took off from my orange landing pad. I had forgotten to change out the battery so after just a few minutes I got the "battery level low" warning so I engaged RTH. I got a message I have never seen before after some 2 1/2 years of piloting both my Phantom 3 Advanced and the Mavic Pro. The message said "Return to "current" location or "original" location. Without really thinking it out, I chose "current" location. Aircraft took off and did not return home. Using the log and latitude and longitude coordinates, located the Mavic via Google Earth but it was not reachable. Anyone ever have this message appear? Plan on calling DJI

I had the same experience recently and chose the "current location" option. Luckily for me I happened to just catch sight of it as it headed in a different location from me behind a hill. I quickly checked the map and saw its flight path was not back to me but a previous home point earlier that morning. I immediately halted the flight, put the Magic to 150 feet and used the map to fly it back within sight. One hell of a scary moment but shows how easily it can be lost, so I do feel for you.
However, before setting off on that flight I had checked my phone for the Home Point and it was correctly set so I don't know if this was an error or me not understanding the GPS point in my phone.
Anyway, hope you get your drone back ...

 

[Lone Droner 10]

I had the same experience recently and chose the "current location" option. Luckily for me I happened to just catch sight of it as it headed in a different location from me behind a hill. I quickly checked the map and saw its flight path was not back to me but a previous home point earlier that morning. I immediately halted the flight, put the Magic to 150 feet and used the map to fly it back within sight. One **** of a scary moment but shows how easily it can be lost, so I do feel for you.
However, before setting off on that flight I had checked my phone for the Home Point and it was correctly set so I don't know if this was an error or me not understanding the GPS point in my phone.
Anyway, hope you get your drone back ...

Drone is lost and I have sent my flight records to DJI for review. In reading posts since I lost it on 8/10, others have had similar problems but, like you, were able to return the Mavic safely. DJI is not getting real good press regarding their customer service so I am waiting to read what DJI determines caused the "fly away" and RTH failure. The Mavic was less than one week old!

 

[scjerry]

OK - to summarize the problem: When you dynamically set the home point to the RC location, then the aircraft sometimes heads to a different location on RTH. Based on more than one test (how many?) you see a correlation between that behavior and whether or not you just calibrated the compass. Is that correct?

Your hypothesis is that there is a connection between the compass calibration process and the ability of the aircraft to store a home point - correct?

In your screen shots the recorded home point is clearly in the region of 400 m south east of your position, based on the distance readout. That distance should have been what you checked on the reset, and would have been an immediate warning that the the mobile device had supplied a spurious location.

Given that the home point has no compass calibration data, or even bearing data, related to it - it's just a position - why would you not hypothesize first that the cause of the problem is a bad GPS fix on your mobile device, since that is where the dynamic home point position data are coming from? That's a much simpler explanation that you have not ruled out and that does not require some extraordinary mess up in the programming and internal data structures in the FC. My hypothesis would be that the correlation you saw a couple of times with compass calibration is just coincidence.

As pointed out by others, without the flight log or, preferably, the DAT file, it is impossible to be completely definitive, although this is such a relatively common problem (bad initial home point on the aircraft or bad dynamic home point data from the mobile device) that the answer looks pretty clear to me. Take a look over on PhantomPilots to see how many times this has come up in the past.

sar104
Question before I proceed further. "Should invocation of Master Control -> Home Point to Me (little man icon), result in a new HP, thus overriding the launch HP?"

 

[woodsglen]

WOW, what an eye-opening discussion for a MP pilot with very few hours - mostly within radius of maybe 300 yards and alt 365 feet.
I use an ipad, not internet, no mobile data.

So my question to the experienced group is: When my MP asks if i want to update my "home point", does this mean that the sat/gps data coming to the mavic is being sent to my Ipad?
I appreciate all of the learning from the forum since I started having time to fly.
Thanks!
WG

 

[msinger]

When my MP asks if i want to update my "home point", does this mean that the sat/gps data coming to the mavic is being sent to my Ipad?

Yes -- if you're being prompted to choose between the current and original location. You should never see that prompt if using a Wi-Fi only iPad since it does not contain a GPS receiver.

 

[woodsglen]

So, when I hear "Home point updated" - Then I'm not going to experience what happened in the original post. Thanks Msinger!

 

[msinger]

So, when I hear "Home point updated" - Then I'm not going to experience what happened in the original post.

Not exactly. If you don't choose current location at that prompt, then there is no chance you'll experience what happened in the original post.

 

[m0j0]

You don't think calibrating your compass in a new location BEFORE flight is a good idea to potentially prevent a flyaway?
I opened my book again after 10 months of owning my Mavic and Page 52 wasn't relevant to my comment..

Not only do I not think it prevents flyaways I think it far more likely to cause an issue. If you have a solid calibration leave it alone. I can't stress this enough. In 2.5 million feet flown I have calibrated 5 times (4 times with P3s). I have flown on both US coasts and points in between. I have a Mavic that I have never calibrated. It flies perfectly. I have a second that I have calibrated once in a perfectly sanitized area it also flies perfectly. I have never had a compass related incident with a Mavic. I have only had one with a Phantom 3 which was corrected by flying away from the huge metal tower.

Read the calibration thread and follow the advice from that post. Do not calibrate otherwise.