Needs help with calibration

[Ronny St.]

Hi all,
I am new to drones, so sorry if my question seems a bit stupid to you
My mini has a strange thing... The arrow in the middle downpart of the GoFly app screen should give you an indication of where the drone is in relation to the remote controller and in which direction it is heading.
But for some reason it does not work for me as intended. I have to stand in a position so that the drone makes an angle of 270° to get the arrow in the middle and pointing forwards. So, if I make the comparison with a clock, my position has to be at 9 o'clock to get the arrow pointing at 12 o'clock. That is very annoying since it makes it very difficult to know your own position relative to the aircraft and to give it the rigt commands.
I calibrated the compass and the IMU but found it strange that the directions (if one follows the figures displayed) the app gives you is to put the drone first on its left and then on its right side. On youtube I found a movies telling me just the opposite:

at 4:12.
I recalibrated several times and then got a compass calibration error, so I calibrated the compass again.
Nevertheless the arrow seems to live its own life. So I begin wondering it that has to do with the IMU whatsoever.
What I am doing wrong and how can I get the arrow pointing in the right direction and on the right spot of the semi-circle?
Thanks in advance for replying!
Greetings from Belgium,
Ronny

 

[Xanadu]

I do have the same issue, you can look in this thread for Android and in this one for iOs devices.

As @timinator mentioned in that first topic: If you use the map, the orientation of the triangle is a reliable indicator which way the aircraft is pointing.

I found this to be accurate at all times and I have been consulting this to check the AC orientation.

 

[Ronny St.]

Thanks Xanadu for the fast reply! So I am clearly not the only one with that issue. I think DJI should incorporate this be solved in the next update, since it is quite a handy little if it should work. Now it is giving only misleading information and that is confusing and pointless.
PS Just watched the video in the link you gave me and that explains a lot. Now we can only hope that DJI will solve this issue soon. Now that I know what the pointer is called I soon found a longer youtube video on this issue at

 

[old man mavic]

it is something that relies on the compass in the device ,i have ignored it from the very beginning and as stated above i use the drone indicator in the map view to know which way it is flying relative to my position,

 

[Ronny St.]

it is something that relies on the compass in the device ,i have ignored it from the very beginning and as stated above i use the drone indicator in the map view to know which way it is flying relative to my position,

As long as it is not fixed by DJI all users of the Mavic Mini should do that, but should be warned to be aware of the problem.

 

[Xanadu]

As OMM said it's related to the phone's compass but you are right, this should probably be something that DJI should bring to the users attention.

 

[PanX]

had same problem on 1st flight, i did compas calibration (iPhone 5s) and till then is perfect (i believe mostly has to do with the behavior of inertia sensors in your phone).

 

[Frankeh]

Hi all,
I am new to drones, so sorry if my question seems a bit stupid to you
My mini has a strange thing... The arrow in the middle downpart of the GoFly app screen should give you an indication of where the drone is in relation to the remote controller and in which direction it is heading.
But for some reason it does not work for me as intended. I have to stand in a position so that the drone makes an angle of 270° to get the arrow in the middle and pointing forwards. So, if I make the comparison with a clock, my position has to be at 9 o'clock to get the arrow pointing at 12 o'clock. That is very annoying since it makes it very difficult to know your own position relative to the aircraft and to give it the rigt commands.
I calibrated the compass and the IMU but found it strange that the directions (if one follows the figures displayed) the app gives you is to put the drone first on its left and then on its right side. On youtube I found a movies telling me just the opposite:

at 4:12.
I recalibrated several times and then got a compass calibration error, so I calibrated the compass again.
Nevertheless the arrow seems to live its own life. So I begin wondering it that has to do with the IMU whatsoever.
What I am doing wrong and how can I get the arrow pointing in the right direction and on the right spot of the semi-circle?
Thanks in advance for replying!
Greetings from Belgium,
Ronny

I have the same problem and it started with the last software and firmware update. I have calibrated the drone's compass and IMU. After reading these responses I calibrated my phone's compass. Nothing has made any difference.

 

[Ronny St.]

Want to come back in this topic to add more info. I recalibrated the compass of my Samsung Galaxy S8 and that seems to have some positive effect, but not solving the issue completely. I just finished a flight and tried to work with the map. It works but it is much more easier to follow the aircraft it the orientation icon would react correctly and is a bit live driving a car where you keep an eye on the speedometer etc. all the time.
I also noticed that when I choose for 'satellite' or 'mixed' (which in my opinion gives the most easy-to-interprete solution) I get something that looks like a yellowish filter above the displayd map and it is a bit annoying to have a clear vue. I guess that it is due to the fact that I am in an altitude zone, so maybe it is some sort of warning from the DJI software, since the airport of Deurne is only about 10 km from here, and when I scroll further away into the map it disappears.

Update: just did a few other flights today. The first one displayted the orientation icon perfectly! But then in another flight I had my first crash I was trying out the speed setting and flew into a hedge. Dumb of me, I didn't take into account that it needs some distance to stop. Whatever, no visible damage so I shut down the aircraft and the controller and restarted it all. Luckily it worked perfectly, whew. The only strange thing I noticed was that the next flight after the crash the aircraft orientation icon was again 90 degrees off. Strange. Is it possible that due to the thumbling down it needs to be recalibrated?

 

[PanX]

Lets make an experiment, open the phone, then the DJI-FLY, then open the drone, wait to initialise, then stand exactly BEHIND the drone, THEN "Connect ..." and when you see the camera view, start to walk around, what happens with orientation?

 

[Xanadu]

I had tried that before, same thing, orientation is still off. As discussed in other related threads (see post #2), the discrepancy in the orientation seems to be due to the compass of the phone and the interference from the RC to the phone's compass.

 

[Ronny St.]

Just did a experiment, but not exactly as PanX suggested but it was based on it. I mounted the phone in the remote, switched on the remote and looked at the compass: it gave me the message 'Strong magnetic fiels, calibrate you compass'. So I recalibrated my compass while in the remote. After that I switched on the aircraft. It worked perfectly! Maybe that can be the solution
One detail:when I took the phone out of the remote and started the compass, it told me I had to recalibrate it, probably because the 'strong magnetic' field was nog langer there. I did not do that and tomorrow I will try flying again without recalibrating the compass first to see if the phone keeps his compass settings. I will let you know the results!
I also considered the possibility of making a 'magnetic shield ' between the remote and the phone eliminating interference. But I did not yet figured out how to do that.

 

[Ronny St.]

Yesterday it worked perfectly at first and the next flight it was complete off again. I guess it must be possible to add in the next firmware update an extra field in 'options' where one could manually adjust the orientation of the aircraft orientation 'icon'. I find it much more difficult to use the map to follow the aircraft continuously.

 

[Xanadu]

I was about to test what you did and then I forgot about it and got it up in the air. During my usual one minute hoovering checking that everything is ok, the orientation was again a bit off. I ignored it and started flying.

While the MM was far away I wanted to do some RTH tests, so at some point, I initiated RTH, and while it was coming towards me, after cancelling it, when I checked the orientation it was pointing where it should be. The only difference compared to other flights was that at that point in flight, it had 20 satellites locked. After that, it stayed like this for the rest of the flight. I will also be confirming this on my next flight (whether at some point mid-flight, the orientation is corrected and whether it depends on the number of satellites).

 

[Ronny St.]

I was about to test what you did and then I forgot about it and got it up in the air. During my usual one minute hoovering checking that everything is ok, the orientation was again a bit off. I ignored it and started flying.

While the MM was far away I wanted to do some RTH tests, so at some point, I initiated RTH, and while it was coming towards me, after cancelling it, when I checked the orientation it was pointing where it should be. The only difference compared to other flights was that at that point in flight, it had 20 satellites locked. After that, it stayed like this for the rest of the flight. I will also be confirming this on my next flight (whether at some point mid-flight, the orientation is corrected and whether it depends on the number of satellites).

That is an interesting observation! Maybe the number of satellites is the key factor. Pls let us know what you experience in your next flights

 

[PanX]

If it initialise the orientation using satelites perhaps the problem is that we are too close to the craft when it is initialising its position. Lets try take off being 4-5 meters away.

 

[Ronny St.]

If it initialise the orientation using satelites perhaps the problem is that we are too close to the craft when it is initialising its position. Lets try take off being 4-5 meters away.

I am curious what you will find (and I am sure a lot of us are )

 

[Xanadu]

If it initialise the orientation using satelites perhaps the problem is that we are too close to the craft when it is initialising its position. Lets try take off being 4-5 meters away.

Actually, this is what I did yesterday. Powered on the RC and then the AC and moved further back around 5 meters away. When I took off, the orientation was off (just like it was even when I was taking off 1 meter away from it).