Implementing Positioning Algorithms Using Accelerometers AN3397

[yellingdog]

Hi,

I'm reading the application note AN3397 and was wondering in what unit is the measured position in?

In the note I don't see Time being a variable anywhere. It is stated that the time between samples MUST be the same however.

So a is in m/s2

v is in m/s ?

and positionX in what?

Here are the integrations I'm doing, with sampling rate 1.5ms:

velocityx[1]= velocityx[0]+ accelerationx[0]+ ((accelerationx[1] -accelerationx[0])/2)*0.0015;
positionX[1]= positionX[0] + velocityx[0] + ((velocityx[1] - velocityx[0])/2)*0.0015;

Do they seem correct?

thanks

[diazmarin09]

Hello,

You are correct, in this case the time between samples MUST always be the same. Errors can be generated if this time is not equal. I do recommend a faster sampling frequency for accurate results.

So a double integration of your samples (after a double filtering process and calibration) is the requirement needed to obtain the position using the acceleration data.

I hope this information helps.

Regards,

David

[diazmarin09]

Hello,

The position is expressed in m (meters).

Note that in order to create a coherent value, sampling time must always be the same.

I hope this information helps.

Regards,

David

[yellingdog]

Hi David,

So T doesn't have to be included in the calculations as long as it is a constant?

thanks