Page 1 of 1

Tracking Tools + VRPN

Posted: Fri Apr 08, 2011 7:34 am
by araa
Hello,

I'm using Tracking Tools 2.3.1 (Right handed system) & VRPN. I would like to obtain the position of a target point relative to the pivot point of my rigid body. First, I tried using the offsets of Tracking Tools and it worked. Now, I�m using the quaternion and the position of the pivot point (VRPN) but, sometimes, the position of the target point is extremely bad.

The equations which I use are the following ones:

[Xtarget] = R1*[Xpivot]+R2*[Ypivot]+R3[Zpivot]+[XOffset]
[Ytarget] = R4*[Xpivot]+R5*[Ypivot]+R6[Zpivot]+[YOffset]
[Ztarget] = R7*[Xpivot]+R8*[Ypivot]+R9[Zpivot]+[ZOffset]


R1 = 1 - 2*(t.quat[1]*t.quat[1]+t.quat[2]*t.quat[2]);
R2 = 2*(t.quat[0]*t.quat[1]-t.quat[3]*t.quat[2]);
R3 = 2*(t.quat[0]*t.quat[2]+t.quat[3]*t.quat[1]);
R4 = 2*(t.quat[0]*t.quat[1]+t.quat[3]*t.quat[2]);
R5 = 1 - 2*(t.quat[0]*t.quat[0]+t.quat[2]*t.quat[2]);
R6 = 2*(t.quat[1]*t.quat[2]-t.quat[3]*t.quat[0]);
R7 = 2*(t.quat[0]*t.quat[2]-t.quat[3]*t.quat[1]);
R8 = 2*(t.quat[1]*t.quat[2]+t.quat[3]*t.quat[0]);
R9 = 1 - 2*(t.quat[0]*t.quat[0]+t.quat[1]*t.quat[1]);

What's the problem?

Thanks,

araa

Re: Tracking Tools + VRPN

Posted: Fri Apr 08, 2011 10:31 am
by beckdo
I think I understand. You have a point in space that you want in a coordinate system based on the rigid body instead of the world coordinate system.

Re: Tracking Tools + VRPN

Posted: Mon Apr 11, 2011 1:28 am
by araa
Sorry, what I wanted to say is that both the pivot and target points belong to the same rigid body but Tracking Tools (through VRPN) just give the position of the pivot point. I also have the offsets between both points. I would like to obtain the position of the target point in the world coordinate system.

Re: Tracking Tools + VRPN

Posted: Tue Apr 12, 2011 10:16 am
by beckdo
If you want to avoid the math, you can make the target point the pivot point of the rigid body.

Re: Tracking Tools + VRPN

Posted: Wed Apr 13, 2011 7:39 am
by araa
The target point is not a reflective marker. We would like not to use the offsets of Traking Tools.

Currently, we are not sure if the problem of having positions of the target point extremely bad comes from the equations above or from Tracking Tools. When we move the rigid body slowly and we plot the values of each term of the quaternion separately, we have seen some peaks which correspond with extremely bad target point positions. :-(

Re: Tracking Tools + VRPN

Posted: Thu Apr 14, 2011 3:48 pm
by NaturalPoint - Mike
Could you send a CSV of data that exibits these traits to support@naturalpoint.com? If you have specific frame numbers that could greatly help.

Re: Tracking Tools + VRPN

Posted: Tue Oct 04, 2011 11:52 am
by maxlem
I have a similar need. I would like to set the pivot point to a precise position in the rigid body. A very simple solution (software wise) would be to allow to set the pivot point at an offset on a line between two markers.

Example :

- create a tracker with 3 or more markers
- select two points among tracker's markers
- create pivot on the line between the two selected markers