Skip to main content
Visitor II
February 2, 2023
Question

LSM6DSOX AHRS Yaw measurement not correct

  • February 2, 2023
  • 1 reply
  • 1311 views

Hi there!

We are using the LSM6DSOX in our product and doing sensor fusion using xioTechnologies/Fusion.

Our hardware is mass produced and we are using the LSM6DSOX sensor with no magnetometer using 52Hz ODR and 104Hz ODR. The change in ODR doesn't seem to make any difference.

We have calibrated our sensors, especially the gyroscope.

Using xioTechnologies/Fusion algorithm with calibrated sensor reading removes the yaw drift and gives us very stable measurements over time.

0693W00000Y9rNqQAJ.png

What we noticed though, is that yaw rotations are not measured correctly. This happens both with and without calibration.

When rotating the device 180° in real life, the yaw measurement is not ~180°:

- for one device we are missing ~20° --> 180° IRL ≡ ~160° for the device

- for another one, we get ~10° more --> 180° IRL ≡ ~190° for the device

The measures are stable for each device and the error is always ~ the same, but it's a big issue for us to not be able to have our device rotate correctly when we need them to.

🙋 do you have any ideas what could be causing that? and how to fix it?

Thanks a lot for your kind help :)

    This topic has been closed for replies.

    1 reply

    Technical Moderator
    April 12, 2023

    Hi @ladislas​ ,

    There are many parameters that impact on the angle calculation, such as sensitivity, cross axis and sample time.

    When you write "calibrated sensor" you mean cross axis calibration?

    How your algorithm measures time?