What format is input gyroscope and accel? about fusion

[xioTechnologies]

The type of the argument gyroscope is FusionVector. This is a structure with three members: x, y, and z. These are the x, y, and z axis measurements provided by the gyroscope.

Pitch, roll, and yaw are Euler angles. These terms are not applicable to gyroscope measurements of angular velocity. The word "normalized" has a specific meaning when referring to vectors. The gyroscope vector should not be normalised.

[David-OConnor]

Thank you for the detailed explanation!

The IMU gyroscopes I've dealt with measure angular velocity in 3 axes, which can be described a pitch, roll, and yaw. How would you use these in this library's update function API?

typedef union {
    float array[3];

    struct {
        float x;
        float y;
        float z;
    } axis;
} FusionVector;

My best guess is the mapping is x: pitch, y: roll, z: yaw, based on the axes of rotation.

Both the gyro and accel readings are of type FusionVector. This makes sense for the accelerator, since it really is an acceleration vector in 3 axes. It's a bit... muddier for the gyro.

[xioTechnologies]

Gyroscope axes are usually described as: x, y, and z. It sounds like the gyroscope you are using has instead assigned the labels: pitch, roll, and yaw. The correspondence of these labels is ambiguous. Are you able to share a link to the documentation?

[David-OConnor]

Looking at the datasheet, it's described as you say - around the X, Y, and Z axes. The pitch/yaw/roll terminology isn't part of the IMU, but a convention I've been using, where x, y, and z are linear directions, and pitch, roll, and yaw are angular directions. With this in mind, your explanation solves this entirely. Thank you very much!

So, the mapping I'm now using, from an internal struct works out to be:

let gyro_data_for_ahrs = Vec3 {
    x: imu_data.v_pitch,
    y: imu_data.v_roll,
    z: imu_data.v_yaw,
};
let accel_data_for_ahrs = Vec3 {
    x: imu_data.a_x,
    y: imu_data.a_y,
    z: imu_data.a_z,
};

[David-OConnor]

Fixed by #9