Skip to main content
Visitor II
April 7, 2022
Solved

LSM6DSM pre-process data information

  • April 7, 2022
  • 1 reply
  • 5588 views

Dear Sir/Madam,

I am currently working on the LSM6DSM IMU for a visual-inertial SLAM use and I have some questions about its features.

Here is my IMU settings :

Gyro :

ODR = 208Hz

FS = 2000dps

High Pass 260mHz

Acc :

ODR = 208Hz

FS = 2g

Analog BW 400Hz

Low noise Low Pass ODR/100

- Do you have the Noise Density and Random Walk values for the gyroscope and the accelerometer and also their different bias ?

- Do you know if these data are unique for each IMU or all the LSDM6DSM have all the same reference values ?

- Should we pre-process our raw IMU data before using them ?

Thank you in advance.

    This topic has been closed for replies.
    Best answer by Eleon BORLINI

    Hi @RRand.1​ ,

    you might find some data of your interest in this thread.

    These data are intended to be an average on a representative sample of LSM6DSM production lots.

    About data preprocessing, it it usually depends upon the target application. If you want, for example, integrate your data to find velocity and displacement or angle, you have to calibrate the sensor (i.e. reducing the offsets as much as possible, such a the zero G level and the zero rate level) at the beginning of the data processing, and, possibly, at the sensor level (e.g. using the dedicated offset registers such as the X_OFS_USR (73h), see datasheet p.107).

    If my reply answered your question, please click on Select as Best at the bottom of this post. This will help other users with the same issue to find the answer faster. 

    -Eleon

    1 reply

    ST Employee
    April 8, 2022

    Hi @RRand.1​ ,

    you might find some data of your interest in this thread.

    These data are intended to be an average on a representative sample of LSM6DSM production lots.

    About data preprocessing, it it usually depends upon the target application. If you want, for example, integrate your data to find velocity and displacement or angle, you have to calibrate the sensor (i.e. reducing the offsets as much as possible, such a the zero G level and the zero rate level) at the beginning of the data processing, and, possibly, at the sensor level (e.g. using the dedicated offset registers such as the X_OFS_USR (73h), see datasheet p.107).

    If my reply answered your question, please click on Select as Best at the bottom of this post. This will help other users with the same issue to find the answer faster. 

    -Eleon