Wiimote Project
General Wiimote and Technology => General Hardware Talk => Topic started by: Rorschach on November 27, 2009, 06:19:45 AM

Hello,
I'm quite inexperienced when it comes to programming, but I've been using LabVIEW to collect accelerometer and IR data from the wiimote (with the help of the information on this site, http://decibel.ni.com/content/docs/DOC1353 ). I've managed to write the data in spreadsheet files, but now I'm figuring out what I got exactly (and what I can do with it).
Am I right in assuming that,
1.
The accelerometers measure the proper acceleration they experience relative to freefall.
Acceleration is quantified in the SI unit m/s2 or popularly in terms of gforce (g).
The accelerometers measure linear accelerations along three fixed axes; they are unable to distinguish between linear motions and rotations, and so they are capable of measuring only pitch and roll angles?
2.
The IR data I get is either raw or .. well notraw of X and Y. But X and Y of what? I've been searching all over the net but I can't get a definitive answer. Is it the (X,Y) position of the IR LED I'm getting here?
In theory can the IR data help you in obtaining yaw and the position of the wiimote (which the accelerometers can't supply)?
Thank you for your time.

I did some more digging and I'm pretty confident about the following information, but I'd appreciate it if someone truly knowledgeable could confirm it.
IR
The raw X and Y values are the IR Led (center) position on the camera's image plane?
The nonraw or normalized data I get with LabVIEW is just easier to work with I gather.
Accelerometer
The accelerometers measure linear accelerations along three fixed axes; they are unable to distinguish between linear motions and rotations, and so they are capable of measuring only pitch and roll angles.

Hi,
If you are using the Wiimote Lib by Brian Peek, the raw X and Y values starts from the upper left corner (if I remembered correctly) as (0,0) and the opposite corner as (767, 1023).
Therefore, the center of the camera's plane will be middle of the two points.
For the normalized data or the raw data, you can just do a simple relative transformation to shift the axis to the center.
If you are trying to obtain data for more degrees of freedom than just pitch and roll angles, you should try out the Wii Motionplus or use the IR camera to assist with the computation of the orientation.
http://www.brianpeek.com/blog/archive/2009/07/20/wiimotelib18beta1posted.aspx (http://www.brianpeek.com/blog/archive/2009/07/20/wiimotelib18beta1posted.aspx)
Regards,
Boon Jin

Pardon me for not being clear on that, but yes, I am using the Wiimote Lib by Brian Peek.
And so we are talking about the Image Plane of the camera then (I wasn't entirely certain of that). I'm using Labview to help me understand the data better and I think it already blurts out the center of the (IR) points, but I'll look into that.
Thank you for the additional info, it's nice to know if you're on the right track or not.