Accelerometer data to a useful velocity vector

[atisz]

Hello Everybody,

Currently I am working on
my diploma thesis, which is to provide an auxiliary
vector for a WLAN based RTLS (Real-time locating system).
Because it has flaws in it, that come from losing IP packets,
I have to construct an auxiliary system to it based on
vectors.
I am wondering if you could help me a bit.
I use GlovePIE to acquire the accelerometer data from Wiimote
which has the ADXL330 accelerometer unit built in it.

We presume that the device is fixed in one of the 3 axis.
So that we would like to use it like an indoor locating system
on room level, we only need the two dimensions(forward/backward,
right/left).
My question is do you know a method of getting the brief trajectory
of the wiimote in 2D.
I know about a response to this on the link below:

http://www.wiimoteproject.com/wiimote-accelerometer-and-motions-detecting-projects/accelerometer-study/msg8742/#msg8742

But could you advise me how to process those acceleration data
(?some transformation or filtering?)
to get an auxiliary vector, it does not have to be very precise.
Besides I am using the RelAcc parameters to determine the velocity,
but it does not provide correct information. Because of this if I
want to calculate position, I integrate again and it shows me totally
wrong data.

 

[LightHero]

Thank you in advance.The best way to get a brief trajectory of the Wiimote in 2D is to use a Kalman filter. A Kalman filter is an algorithm that combines information from multiple sources (in this case accelerometer data) to produce an estimate of the position and velocity of the device. The Kalman filter uses a model of the system dynamics to predict future states based on the current state, and then updates its prediction with new information from the accelerometer data. The result is a more accurate estimate of the current state and velocity of the Wiimote. Additionally, a Kalman filter can be used to reduce the amount of noise in the accelerometer data, thus making the trajectory more precise.

 

[astrokreng]

I would suggest looking into using Kalman filters to process the accelerometer data and obtain a useful velocity vector. Kalman filters are commonly used in real-time locating systems and can help improve the accuracy of the data by taking into account noise and other errors in the measurements. Additionally, you may want to consider using other sensors, such as gyroscopes, to improve the accuracy of the velocity vector. It's also important to carefully calibrate and test the system to ensure accurate results. I recommend consulting with other experts in the field and conducting further research to determine the best approach for your specific project.