Parrot anafi magnetic perturbations / crash

[Deleted member 3201]

Once I got to it I immediately noticed one of the front right propellers was gone, and I couldn't find it so I'm not sure if it lost that in the air or just threw it into the snow on landing upside down.

First off; you are a true hero and excellent pilots, you really rode it down all the way, in a controller matter.

The right front propeller indeed DID break in flight, I can tell you are 100% correct because of the video you supplied:




also, I have the magnetic error as well nearly all the time now in most concrete or near metal take off locations.
I believe that is because they are more strict now with the sensors.

 

[Gipsz Jakab]

I think it is hopeless to complain to the Parrot about a propeller break because you can’t prove to them that the accident was not a collision with a bird or a large insect, and it’s not a warranty issue.

 

[MustangPhil]

@IronSky1 Your idea of measuring the magnetic field strength or deviation with the compass level application is really interesting and I will try it and compare it with a handheld real compass for hiking (I didn’t know this app and used the sensor tests app that includes the magnetometer check on 3 axis).

The reasons of my doubts on relying on an app are the following :

- First, my phone has a protection that keeps it closed with a magnet embedded in the protection material. Those little magnets are also the root cause of the misleading information when you use the GPS app to guide you on a map, when you see your car moving in an awkward position (like a ‘crab’).
- Even when I remove this phone protection, I still have errors that may be induced by the magnetic fields emitted by the phone itself, the components, the battery...
- Applications do usually work not with only one sensor, but do combine many of them, such as the tilt sensors (gyros...) and accelerometer, along with the GPS to determine the direction of the magnetic field depending on where you are on earth.
In France for example, the magnetic field is not exactly normal to the surface of the ground, it plunges to the ground at approx 20°, as at the North Pole, the direction of the magnetic field is ... at 90° just below your feet.
The use of this combination of sensors and the attached algorithms are smart but difficult to evaluate, where the handheld compass only gives you the direction of the magnetic flow... whatever the situation.

I’ll try it !

@Test2000 ... I don’t know witch of the 2 propositions is the most accurate :
- the OP is a hero because he could land his drone in a controlled manner after the loss of a prop (congratulations for you to capture the image of the shadow of the Anafi with the missing prop !)
- The development team at Parrot that could produce an algorithm that lets the drone equilibrate dispite the loss of a prop and land without damage (in a snow-foam blanket)... is the real hero ;o)

All in all, they are still working on the magnetic perturbations algorithm as they mentioned in the dev forum... and it still needs fine tuning...
But it’s a tough subject (they expect to make it work indoors also and I’m very interested for inspection purposes, where I remarked the Anafi is superior to a Phntom 4 for example)

 

[Deleted member 3201]

@IronSky1 Your idea of measuring the magnetic field strength or deviation with the compass level application is really interesting and I will try it and compare it with a handheld real compass for hiking (I didn’t know this app and used the sensor tests app that includes the magnetometer check on 3 axis).

The reasons of my doubts on relying on an app are the following :

- First, my phone has a protection that keeps it closed with a magnet embedded in the protection material. Those little magnets are also the root cause of the misleading information when you use the GPS app to guide you on a map, when you see your car moving in an awkward position (like a ‘crab’).
- Even when I remove this phone protection, I still have errors that may be induced by the magnetic fields emitted by the phone itself, the components, the battery...
- Applications do usually work not with only one sensor, but do combine many of them, such as the tilt sensors (gyros...) and accelerometer, along with the GPS to determine the direction of the magnetic field depending on where you are on earth.
In France for example, the magnetic field is not exactly normal to the surface of the ground, it plunges to the ground at approx 20°, as at the North Pole, the direction of the magnetic field is ... at 90° just below your feet.
The use of this combination of sensors and the attached algorithms are smart but difficult to evaluate, where the handheld compass only gives you the direction of the magnetic flow... whatever the situation.

I’ll try it !

the iPhone compass is only accurate if you use GPS.
Also my iPhone 12 Pro has a strong ring of actual magnets on the back of the phone, but the compass works still, maybe it was calibrated in the factory or the sensor is far away

 

[MustangPhil]

The calibration of the magnetometer a device surrounded by fixed perturbations is the ‘hard iron effect’ and is now well handled by developers... as long as to calibrate your phone magnetometer with the ‘figure 8 dance’ on your phone or the Anafi ‘calibration’.

The problem may come from other factors such as the ‘soft iron effect’ (metallic things or fields that were not present at the time of calibration, your battery level and if your phone uses lots of current at the time of analysis, screen light dimmed or not, you wear an Apple Watch that you did not have during calibration, your mother in law is far to close for confort...)

 

[Deleted member 3201]

@Test2000 ... I don’t know witch of the 2 propositions is the most accurate :
- the OP is a hero because he could land his drone in a controlled manner after the loss of a prop (congratulations for you to capture the image of the shadow of the Anafi with the missing prop !)
- The development team at Parrot that could produce an algorithm that lets the drone equilibrate dispite the loss of a prop and land without damage (in a snow-foam blanket)... is the real hero ;o)

All in all, they are still working on the magnetic perturbations algorithm as they mentioned in the dev forum... and it still needs fine tuning...
But it’s a tough subject (they expect to make it work indoors also and I’m very interested for inspection purposes, where I remarked the Anafi is superior to a Phntom 4 for example)

It is simple basically; you have 4 motors and 1 has only 50% of the props meaning lift cut in half at least for that one. If the other 3 lower their speed then it should be able to hover at least. But I don't know if the software could detect that?


In theory: if you initiate a landing and fly diagonal in the direction opposite of the broken prop it should be flying relatively stable relative to the ground position, I thought that is what the pilot did; combined with gimbal down it indicated to me that he was relatively in control

 

[ParrotAnafiFan]

Is anyone traveling far with their anafi. It could be something your car is passing on the way to the field that effects your anafi magnetically a great deal. Also many electric cars have electric big motors
(and large magnets) in them today. You also may pass under normal high voltage lines on the way. Once they have that much field hitting them is one calibration good enough to solve the issue? Are two or more calibrations better in this scenario? How many feet is too close to a powerline before you run into perturbation issues? A cell tower?

Does it still effect the anafi being off while traveling?