search

kriswiner / MPU9250

Arduino sketches for MPU9250 9DoF with AHRS sensor fusion
1.03k stars 471 forks source link

Altitute / lifting affects Yaw in AHRS not in IMU #214

Open ben5en opened 6 years ago

ben5en commented 6 years ago

Hello!

great thanks to Kris for that fantastic source and all your work!

My question deals more with the quaternion filter: During the implementation of the mpu9250AHRS I noticed a strange behavior: Lifting the board (I am using an evaluation-board with a mpu9250-evaluation board on top) horizontal with no rotation at all affects the yaw value clearly. This only happens when using the AHRS function, while using only the IMU function this effect is not visible.

`MahonyAHRS( ax, -ay, -az, \ gx, -gy, -gz, \ my, -mx, mz);

MahonyIMU(ax, -ay, -az, \ gx, -gy, -gz);`

How is this phenomenon explained? All sensors are calibrated. -> The IMU function is related to Paul Stoffregens source

kriswiner commented 6 years ago

No idea, in general I would expect the 9 dof solution to be better, more stanle. Sounds like cross axis effect. Are you sure the sensor data is in the correct order?

On Tue, Dec 12, 2017 at 4:04 PM ben5en notifications@github.com wrote:

Hello!

great thanks to Kris for that fantastic source and all your work!

My question deals more with the quaternion filter: During the implementation of the mpu9250AHRS I noticed a strange behavior: Lifting the board (I am using an evaluation-board with a mpu9250-evaluation board on top) horizontal with no rotation at all affects the yaw value clearly. This only happens when using the AHRS function, while using only the IMU function this effect is not visible.

`MahonyAHRS( ax, -ay, -az, gx, -gy, -gz, my, -mx, mz);

MahonyIMU(ax, -ay, -az, gx, -gy, -gz);`

How is this phenomenon explained? All sensors are calibrated. -> The IMU function is related to Paul Stoffregens source

— You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub https://github.com/kriswiner/MPU9250/issues/214, or mute the thread https://github.com/notifications/unsubscribe-auth/AGY1quMB-l-Nfl-JB0zB6nxveQ0hipUxks5s_xSDgaJpZM4Q_0YG .

ben5en commented 6 years ago

I tried a lot of variants. I started with your from the MPU9250_M5637_HARS_t3 sketch. I also tripple checked my code. I can not find the mistake and will now build up another board with the MPU9250. Maybe the chip really has a bug?!

ben5en commented 6 years ago

SOLVED: After a lot of code testing it seems like something was disturbing the magnetic field. Using a long USB-Kabel and testing it far away from the computer / monitor works!

Thanks!

kriswiner commented 6 years ago

Yes

On Wed, Dec 13, 2017 at 1:59 PM ben5en notifications@github.com wrote:

SOLVED: After a lot of code testing it seems like something was disturbing the magnetic field. Using a long USB-Kabel and testing it far away from the computer / monitor works!

Have you ever having trouble with magnetic distortion? Thanks!

— You are receiving this because you commented.

Reply to this email directly, view it on GitHub https://github.com/kriswiner/MPU9250/issues/214#issuecomment-351538666, or mute the thread https://github.com/notifications/unsubscribe-auth/AGY1qrU-_FNAahs10k_TAnUkK2g7rmCeks5tAEjAgaJpZM4Q_0YG .