recording a track : what is the precicion I can expect ?

Hey Ronald, thanks for stopping by.

A 10 degrees of measurement IMU (accelerometer + gyroscope +
magnetometer + barometer) like FreeIMU v0.3.5 and 0.4 is capable of
sensing its orientation (yaw, pitch and roll) and altitude while can
only give not very good information about 3d displacement. This is
because the only sensor you can use to do so is the accelerometer but
you would have to double integrate it's data causing propagation of
errors and noise. Can't say if this is accettable in your application.

For starting documenting on these topics, I suggest watching this
video http://www.youtube.com/watch?v=C7JQ7Rpwn2k .. you may also find
my MoS thesis pretty useful
http://www.varesano.net/blog/fabio/my-mos-thesis-using-arduino-tangible-human-computer-interaction

Hope this helps,
Take care,

Fabio

On 11/06/2011 02:01 PM, Ronald van Aalst wrote:
> New question #177751 on FreeIMU:
> https://answers.launchpad.net/freeimu/+question/177751
>
> Hi,
>
> This FreeIMU seems to be what I need.
> for an imaging task, I need to know the exact trajectory of a sensor, so I can compensate the
> received signals (and act as if the signals were received while moving in a straight line).
>
> So, I need to determine (after the fact) what the error of the actual trajectory was, with regard to the straight line
> between start and finish (for a distance of about one to two meters).
> the accuracy should be less than half a wavelength, so in my case in the order of a few millimeters.
>
> what accuracy can I expect from the current FreeIMU ? could this work or is it an unreachable goal from the beginning ?
>
> I have not yet dived into the technicalities of FreeIMU and IMU tech in general, so if there is a short explanation
> about the factors defining accuracy around (or a good starting point for reading about this technology)
> I would be most grateful for some pointers.
>
> Regards,
> Ronald
>