DJI Mavic, Air and Mini Drones
Friendly, Helpful & Knowledgeable Community
Join Us Now

general Compass question

BerndM

Well-Known Member
Joined
Jan 8, 2017
Messages
895
Reactions
464
Location
Planet Earth
I have never understood the need for the compasses in any of my GPS equipped quads. I understand perfectly that a GPS has no compass heading orientation WHILE HOVERING, but the instant the quad starts moving, the compass heading is absolutely and perfectly established. So why all the hassle with compass calibration?? I just don't get it. It seems totally redundant and needlessly complicated.
 
Winds can apply yaw torque too. So to confirm heading you would have to *constantly* be testing heading by *constantly* moving and comparing GPS and gyro readings. To confirm direction with GPS you would need to travel well over a meter each time you needed confirmation.
 
I'm sure future consumer drones won't require the compass dance or be affected by magnetic interference. The new M-200 series quad by DJI has stronger resistance to magnetic interference and cm precision. I'm sure that will bleed down to their consumer drones. Hopefully it'll bleed down to Mavic 2. It'll be the edge that keeps DJI at #1. Not that they are in any danger of losing that spot.
 
It takes a fair amount of movement to correctly determine heading with GPS, back a few years ago satnavs in cars struggled to determine orientation if stationary. With a camera on the front of a UAV you want that to remain on your subject even when stationary so the compass is pretty important as even when hovering the UAV is countering the elements to maintain position and heading.
 
There is a difference between heading and track. GPS does not provide heading. Heading is the angle from magnetic north that the aircraft points, whether moving or not, and is only available from a compass. Add in declination and you can get true heading. Track is the direction that an aircraft has traveled over the ground (what is behind you). Course, on the other hand, is the direction that you are going. GPS can provide track by calculating the difference between two position fixes. Comparing track with heading will give you drift. This along with groundspeed (also available from GPS) will give you wind speed and direction, if you know your airspeed. Heading, track, and course have very specific meanings and uses in aircraft navigation.
 
  • Like
Reactions: halley
One additional term that I forgot to mention - magnetic deviation (not declination) is the error in the compass reading due to nearby metallic objects. It is applied as a correction to magnetic compass heading. Deviation is computed by calibrating the compass. Declination is the angular difference between magnetic north and true north. It varies based on position on the earth's surface. It also varies over time.
 
so I know this is a bit off topic but under the sensor section on the app when you pull up the compass or IMU signals it shows you the signal strength. Green is obviously good but do you want to see more green on the horizontal bars or less green? All of mine are green but have very little bar colored in.....so is more or less better?
 
One additional term that I forgot to mention - magnetic deviation (not declination) is the error in the compass reading due to nearby metallic objects. It is applied as a correction to magnetic compass heading. Deviation is computed by calibrating the compass. Declination is the angular difference between magnetic north and true north. It varies based on position on the earth's surface. It also varies over time.
Deviation (or declination) isn't determined during a compass calibration. It can't be. Determining deviation would require the AC be placed in a stationary position with a known heading and then the AC would have to be told what the heading was.
 
Deviation (or declination) isn't determined during a compass calibration. It can't be. Determining deviation would require the AC be placed in a stationary position with a known heading and then the AC would have to be told what the heading was.
The compass calibration does indeed compute the deviation, as explained here by msinger -
Compass Calibration Guide

If you're interested, here is the relevant chapter from Bowditch on correcting for compass errors.
http://fer3.com/arc/imgx/bowditch1995/chapt06.pdf
Granted, this is for ships at sea, but the same principles apply to aircraft navigation. For aircraft, this calibration is called "swinging the compass".
 
so I know this is a bit off topic but under the sensor section on the app when you pull up the compass or IMU signals it shows you the signal strength. Green is obviously good but do you want to see more green on the horizontal bars or less green? All of mine are green but have very little bar colored in.....so is more or less better?
Less is better, the more green the more interference by local anomaly is blocking your compass.
 
OP said "the instant the quad starts moving, the compass heading is absolutely and perfectly established".

Yes, true. The known COURSE of the vehicle is now known at this point, but is it moving forwards, sideways, backwards relative to the body of the quad? The compass is needed to determine the orientation of the quad relative to its course, otherwise flight controller corrections are impossible. Now, that said- the IMU can detect movement based on the gyros and attempt to calculate what it thinks is the orientation of the quad relative to course direction, but that is complex and subject to accumulation errors. I suspect the Mavic indeed has these calculations, but the compass is still an integral sensor to resolving what is actually going on.

Airplanes rarely fly backwards or sideways, quads do. Compass is essential for quads. Not so much for airplanes that are not capable of flying backwards or sideways.
 
Less is better, the more green the more interference by local anomaly is blocking your compass.
Thanks, I figured this out as well after posting. Since I have never seen anything other than green (excellent) I didn't know that as the bar begins to go to the right it will eventually turn yellow and then red. It is just opposite of what I thought when comparing it to the signal strength of say my Dish receiver where the more signal strength you have the more the bar lights up....
 
The compass calibration does indeed compute the deviation, as explained here by msinger -
Compass Calibration Guide

If you're interested, here is the relevant chapter from Bowditch on correcting for compass errors.
http://fer3.com/arc/imgx/bowditch1995/chapt06.pdf
Granted, this is for ships at sea, but the same principles apply to aircraft navigation. For aircraft, this calibration is called "swinging the compass".
You're right. I've seen the term deviation used here in this forum and the Phantom forums to refer to local geomagnetic distortions, i.e. not declination but distortions that are external to the AC and not caused by the AC itself. The paper you referenced clearly defines deviation as being caused by on board stuff. The whole point of the compass calibration is to compensate for the deviations.
 

DJI Drone Deals

New Threads

Forum statistics

Threads
130,600
Messages
1,554,279
Members
159,607
Latest member
Schmidteh121