05-14-2014 08:58 AM
Hi,
I'm using 3-axis accelerometer, and i want to calculate velocity and distance by integration.
I'm trying to filter the noise that comes with the signal.
The velocity integration is fine. I'm getting velocity that is very close to zero when my robot is standing still.
However, the second integration is diverging because of the value of velocity.
How can I filter the siganl to get distance that is not diverging?
My vi file is attached.
Thank You.
05-14-2014 12:03 PM
Please write the model used accelerometer or example of raw-data array. Triaxial accelerometer If, then calculate the distance traveled should be carried out based on the total vector whose projections on each axis accelerometer and outputs.
05-15-2014 09:45 AM
sagishy,
how far does your distance value diverge? and i second pavel. could you share your raw data?
05-15-2014 01:12 PM
The divergence you see may be the result of a constant offset from the first integration. The integral of Cdt is Ct. If you remove the mean after the first integration, the Ct term will be removed from the second integration. You do need to be careful that you do not remove a slowly varying or constant term which is actually part of your real data.
Lynn