0
$\begingroup$

I am new to this platform so I apologize if I am asking question in an incorrect manner

I am not a physic student hence the reason I am here to clarify some of my doubts. I have this idea of using an accelerometer to just see if some object (e.g. toy car, bicycle) is moving or not through the acceleration data by the accelerometer.

I know accelerometer does not give linear acceleration but with gravity hence will compensate the gravity to obtain the linear acceleration which is $(0,0,0)$ for $x, y, z$.

Using the linear acceleration of all axis i will find the total acceleration and then integrate it to find the velocity that will prompt me to know if the object is moving if it is $0$ (stationary)= not moving, if it is above a value it is consider moving.

I just want to know if this method will work, if I am wrong. I would like to seek some advises to do it the correct way

$\endgroup$
3
  • $\begingroup$ This is possible in theory but difficult in practice due to noise and other sources of inaccuracy, and there's no way simple with only an accelerometer to ever know an absolute velocity because you never have a reference. You could, for example, get a good estimate of a change in velocity if the time interval is short, but anything past that becomes difficult. You might want to look into "dead reckoning" to get a sense of the issues. $\endgroup$
    – tom10
    Commented Mar 10, 2021 at 1:28
  • $\begingroup$ Yes it will work, though as tom10 says it may not be very accurate. What you suggest is how inertial navigation systems work. $\endgroup$ Commented Mar 10, 2021 at 11:59
  • $\begingroup$ hi tom 10 and John Rennie! Thank for your reply! I will look more into the resources you guys have provided! $\endgroup$
    – Lxx lxts
    Commented Mar 11, 2021 at 1:53

0

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Browse other questions tagged or ask your own question.