unstable YAW value

[hicode-and]

Hi, first thx a lot for you work ! I have a little issue : i want to observe the yaw (not the absolute because its impossible without a magn). I tried your code V6.12, it work's fine but when i move the sensor i have a new angle (wich seem good) and 10 second after it change, it take +/- 0.60. And i have a permanent drift ( i think 0.40° per min). Have you some solution for that ? thanks

and i use a MPU9250 and a arduino nano !

[fralbo]

Hello, Pitch and Roll drifts are solved using the gravity whatever you use quaternion or complementary filter. Of course, as Yaw axis is colinear with z axis, you cannot use the same method. The only way to suppress the Yaw drift is to connect a magnetometer and use it in the same way as you use gravity for Pitch and Roll.