compass accuracy Phantom 4 advance/pro

[Andy9]

The answer to a question here can't be more exact as the subject is. And It is such since people invented the compass.
The explanation has been done very clear and in detail thanks to 'solentlife' and 'meta4'.

 

[Dor]

Guys,
Thank you very much for all the info.
So if i can sum it up, if I am telling the drone to take picture with heading "000" with litchi or DJI GS PRO or other app that can do so, I can be sure that the heading is reference to the "true north", and there is no specific accuracy number that I can trust on, but we can assume that the tolerance is few degrees, if the compass and imu are calibrated properly and there are not any interference at the site .
If i am wrong with my assumption please correct me

 

[sar104]

Sorry but wording it in technical sounding jargon does not change the fact that YAW is purely a horizontal rotation and only has a numerical factor when referenced.
Your last post now uses the the term YAW VALUE .... which agrees with this.

I'm puzzled. You continually tout your qualifications in these fields, and yet when confronted with the technical definition of heading, which is what DJI aircraft call yaw, you call it "technical sounding jargon".

YAW is a indeterminate horizontal rotation

No. In this context yaw is an Euler angle - the Tait-Bryan z-axis rotation angle, to be precise.

YAW VALUE or number of degrees YAW is a YAW with a numerical addition.

As to being referenced to True North. I agree that the Flight Controller via IMU and Compass, arrives at an approximate True North reference to give a degrees value to its heading.
But Yaw value in fact can be based on any value of the 360.

But it isn't - it's referenced to true north. Why are you continuing to argue about this?

??

Lets take a P3 / P4 that is heading 000 (North) and yaws -30 ... it now has applied a YAW VALUE of -30 and is heading 330.
Now lest assume the P3 / P4 is now heading 010 ? The YAW VALUE that was applied was +40 ...
Take any heading and cause YAW ... read of the new heading and the difference of the two is YAW VALUE.

Where is True North in that ? The only function of North is to provide a reference 000 for direction. If I said to you - I YAW'd 60 degrees ... are you then going to tell me my new heading is 060 ? Even though I never told you what my original heading was ?

This is pure obfuscation, with a nice dose of straw man thrown in. If you wish to insist that "yaw" is a verb, even though it's been a rotation geometry parameter used in aerospace for decades, then this discussion is pointless.

The OP asked what accuracy and how to FLY specific headings ... which is only possible to an approximate accuracy in terms of actual True North, regardless of how many decimals flight log shows.

Which is why the answer, already provided before you decided to try to confuse everything, is that the absolute compass accuracy is not known.

 

[Meta4]

So if i can sum it up, if I am telling the drone to take picture with heading "000" with litchi or DJI GS PRO or other app that can do so, I can be sure that the heading is reference to the "true north", and there is no specific accuracy number that I can trust on, but we can assume that the tolerance is few degrees, if the compass and imu are calibrated properly and there are not any interference at the site .

The simplest way to do what the client is asking for without worrying about the drone's compass, would be to check the bearing specified by the client and see where that is pointing to towards the horizon.
Then launch and put that point in the centre of your frame and shoot away.
WIth the Phantom's wideangle lens you are going to take in 36° either side of the centre point, so as well as the target point, you'll get a lot of extra in the shot anyway..

 

[bsartist]

I'm puzzled. You continually tout your qualifications in these fields, and yet when confronted with the technical definition of heading, which is what DJI aircraft call yaw, you call it "technical sounding jargon".



No. In this context yaw is an Euler angle - the Tait-Bryan z-axis rotation angle, to be precise.



But it isn't - it's referenced to true north. Why are you continuing to argue about this?



This is pure obfuscation, with a nice dose of straw man thrown in. If you wish to insist that "yaw" is a verb, even though it's been a rotation geometry parameter used in aerospace for decades, then this discussion is pointless.



Which is why the answer, already provided before you decided to try to confuse everything, is that the absolute compass accuracy is not known.

 

[solentlife]

Guys,
Thank you very much for all the info.
So if i can sum it up, if I am telling the drone to take picture with heading "000" with litchi or DJI GS PRO or other app that can do so, I can be sure that the heading is reference to the "true north", and there is no specific accuracy number that I can trust on, but we can assume that the tolerance is few degrees, if the compass and imu are calibrated properly and there are not any interference at the site .
If i am wrong with my assumption please correct me

Correct.

 

[solentlife]

I'm puzzled. You continually tout your qualifications in these fields, and yet when confronted with the technical definition of heading, which is what DJI aircraft call yaw, you call it "technical sounding jargon".



No. In this context yaw is an Euler angle - the Tait-Bryan z-axis rotation angle, to be precise.



But it isn't - it's referenced to true north. Why are you continuing to argue about this?



This is pure obfuscation, with a nice dose of straw man thrown in. If you wish to insist that "yaw" is a verb, even though it's been a rotation geometry parameter used in aerospace for decades, then this discussion is pointless.



Which is why the answer, already provided before you decided to try to confuse everything, is that the absolute compass accuracy is not known.

I shall refer you to Nasa Aernautics :

Aircraft Yaw Motion

The FreeDictionary :

yaw

Even that bastion of the online world ... Wiki :

Aircraft principal axes - Wikipedia

With regard to Euler and Tait-Bryan ....

Euler angles - Wikipedia

Why bring in 3 dimensional matters to a single dimension yaw subject ? The Tait-Bryan you MISQUOTE in fact in terms of yaw ... "A yaw will obtain the bearing " ... bearing in this context is the resultant bearing + or - from original heading ... as I described in my previous. In other words - the difference of the two headings

Yaw has no specific reference other than the original heading before the aircraft yawed ...

There is of course another version of YAW ... which is YAWN ... something that your misunderstanding of yaw is causing. As to DJI using term YAW ... as the term for heading .... please provide the evidence that DJI actually do as you say here :>

... technical definition of heading, which is what DJI aircraft call yaw,...

Yaw for all time on boats, aircraft has been the horizontal rotation in the single dimension about a vertical axis.

In fact Yaw has NOTHING whatsoever to do with compass ....

Yawn Yawn ...

As another said - Meta4 and I answered the OP's question and this addition started by your misrepresentation of Yaw is irrelevant ... but interesting !

 

[Tenisyn]

It sounds like you have misunderstood the meaning of the yaw value used by the flight controller. It's the angle between the horizontal projection of the aircraft x-axis and the direction of true north. It's therefore the aircraft's true heading.

I feel there is an important distinction that is being missed. Heading vs. true course. For fixed wing aircraft, they’re essentially one in the same as the direction of travel will always match the direction the nose is facing. In unmanned aircraft, there can be and often is distinct separation between direction of heading and true course.

 

[N017RW]

I feel there is an important distinction that is being missed. Heading vs. true course. For fixed wing aircraft, they’re essentially one in the same as the direction of travel will always match the direction the nose is facing. In unmanned aircraft, there can be and often is distinct separation between direction of heading and true course.

Huh?
Do you have any piloting experience?

Course and heading are often different and that difference is the WCA or wind correction angle.

 

[sar104]

I shall refer you to Nasa Aernautics :

Aircraft Yaw Motion

The FreeDictionary :

yaw

Even that bastion of the online world ... Wiki :

Aircraft principal axes - Wikipedia

With regard to Euler and Tait-Bryan ....

Euler angles - Wikipedia

Why bring in 3 dimensional matters to a single dimension yaw subject ? The Tait-Bryan you MISQUOTE in fact in terms of yaw ... "A yaw will obtain the bearing " ... bearing in this context is the resultant bearing + or - from original heading ... as I described in my previous. In other words - the difference of the two headings

Yaw has no specific reference other than the original heading before the aircraft yawed ...

There is of course another version of YAW ... which is YAWN ... something that your misunderstanding of yaw is causing. As to DJI using term YAW ... as the term for heading .... please provide the evidence that DJI actually do as you say here :>



Yaw for all time on boats, aircraft has been the horizontal rotation in the single dimension about a vertical axis.

In fact Yaw has NOTHING whatsoever to do with compass ....

Yawn Yawn ...

As another said - Meta4 and I answered the OP's question and this addition started by your misrepresentation of Yaw is irrelevant ... but interesting !

I realize that, this being the internet, you can endlessly reply with ever increasing levels of obfuscation to try to save face, so I'm not going to indulge you further. Yawning doesn't improve your argument either - it just makes it more obvious that you don't actually have a point to argue. And sorry, but it's not my job to provide you with evidence of anything. You should probably take the same approach here that you do for GPS questions and follow Mark Twain's sage advice.

 

[sar104]

I feel there is an important distinction that is being missed. Heading vs. true course. For fixed wing aircraft, they’re essentially one in the same as the direction of travel will always match the direction the nose is facing. In unmanned aircraft, there can be and often is distinct separation between direction of heading and true course.

Yes - this discussion is purely about true heading. Actually it's really about gimbal true heading, since the gimbal often has an offset from the aircraft heading. Course is irrelevant to the problem - the aircraft can be flying sideways or backwards or be completely stationary.

 

[solentlife]

I realize that, this being the internet, you can endlessly reply with ever increasing levels of obfuscation to try to save face, so I'm not going to indulge you further. Yawning doesn't improve your argument either - it just makes it more obvious that you don't actually have a point to argue. And sorry, but it's not my job to provide you with evidence of anything. You should probably take the same approach here that you do for GPS questions and follow Mark Twain's sage advice.

Admit you are suffering Pride and wrong ... and it all goes away ...

Yaw will ALWAYS be horizontal rotation about a vertical axis ... how you can honestly keep denying the fact is truly amazing.

No saving face at all .... you made a statement that was fundamental in error. You continue to try and support that error by using irrelevant references. When caught out again - you continue by calling me personally into question.
Trouble is there are new people who come into this hobby and they get misled by posts like yours on yaw ...

The reason you don't provide evidence ... is I already blew your Euler Angle and Tait Bryan out as they are for 3 dimensional models ... and not to singular.

I'm glad you will not 'indulge me further' as that saves the further ridicule of your ridiculous definition of Yaw.

Shall we just remind ourselves what it was another stated that started this ? Yaw is based on True North.

Internet .... yep ...

 

[grego122]

Darn good entertainment! I'm pointed north west sort of more or less depending on who you ask of course!?

 

[Fvunen]

Not pertaining to accuracy, yaw and all the other angles thrown in this discussion, is there anybody with a suggestion on how to blend compass heading in a picture? Following the suggestions made above that the lens on the P4 is so wide that a few degrees left or right from the target don't matter much, it would still be nice to see in which direction you are looking. In particular in areas with few discrete or obvious recognizable points this would be helpful.

 

[solentlife]

Not pertaining to accuracy, yaw and all the other angles thrown in this discussion, is there anybody with a suggestion on how to blend compass heading in a picture? Following the suggestions made above that the lens on the P4 is so wide that a few degrees left or right from the target don't matter much, it would still be nice to see in which direction you are looking. In particular in areas with few discrete or obvious recognizable points this would be helpful.

You can set Go4 to record a subtitle file that tallies with your video. I'm not sure if it can also do it for a photograph.
Using a program such as Handbrake - that subtitle data can be incorporated into video after any edits you wish to make. If left as recorded - its a separate file with same name as the video ... and needs to be in same directory.

You can edit the file in Notepad or other similar plain text editor ... but you must leave the timeline alone to ensure correct timing with video.

If you combine with Handbrake - you could take snapshots from the video - which would then have the data shown ... If you match the snapshot to any photo you take - you can superimpose the data yourself ?

 

[bsartist]

You can set Go4 to record a subtitle file that tallies with your video. I'm not sure if it can also do it for a photograph.
Using a program such as Handbrake - that subtitle data can be incorporated into video after any edits you wish to make. If left as recorded - its a separate file with same name as the video ... and needs to be in same directory.

You can edit the file in Notepad or other similar plain text editor ... but you must leave the timeline alone to ensure correct timing with video.

If you combine with Handbrake - you could take snapshots from the video - which would then have the data shown ... If you match the snapshot to any photo you take - you can superimpose the data yourself ?

I thought GO4 now embeds Irvin the video and not a separate srt file. At least it seems to on my Phamtom 4 Pro (V1). Unless I missed a setting.

 

[solentlife]

I thought GO4 now embeds Irvin the video and not a separate srt file. At least it seems to on my Phamtom 4 Pro (V1). Unless I missed a setting.

Maybe ... I don't use Go4 ...

I assumed it would do similar to Go before 4 where it creates the subtitle file separately.

If it embeds direct ... can you set what data it embeds or does it just put all ? With Go before4 - you can edit it before embedding.

 

[bsartist]

Maybe ... I don't use Go4 ...

I assumed it would do similar to Go before 4 where it creates the subtitle file separately.

If it embeds direct ... can you set what data it embeds or does it just put all ? With Go before4 - you can edit it before embedding.

Yeah on my P3S I could edit the SRT files. With the P4P I haven’t discovered a way to edit that embedded data. Although I haven’t had the need to do so so I haven’t researched a way to extract it.

 

[solentlife]

Yeah on my P3S I could edit the SRT files. With the P4P I haven’t discovered a way to edit that embedded data. Although I haven’t had the need to do so so I haven’t researched a way to extract it.

Tks ... I may have a look see if there is a way ...

The separate file may be a pain in some ways - but being able to easily edit and them embed is in my view a better way to go.