Re: Can I use accelerometer to get position?

Started by t_p_...@yahoo.com June 17, 2008
>i use accelerometer to detect position to.
>1. Mr. tom becker : "what do you mean with "Dead Reckoning"?
>2. Mr.Sam...I have also used the double integration :
>u said that :
>V(k)=V(k-1)+A(k)*DT
>P(k)=P(k-1)+1/2*A(k)*DT*DT
>
>i have the same one for V(k), but for the P(k)=P(k-1)+V(k)*DT. Is it the same??
>3. i dont know why the voltage output in passive condition (not moving), is always change/unstable??is it noise?? do i have to add filter??
>
>4. when i try it, to measure the distance...sometime it gives different results...with fast/medium/low acceleration..whereas it should be the same distance...however the acceleration.
>thanks
>
>
I have worked with analyzation of accelerometer signals and also made simulations and analysis of the signals. As I imagine the issue will be to avoid offset and/or low resolution in the acceleration signal, because when integrating these errors will accumulate into an erroneous velocity which is then integrated into an even more erroneous position. If you insist on using aceleration, your system should be able to calibrate itself frequently.