Angular velocity integration

Please don't post Bullet support questions here, use the above forums instead.
Post Reply
mikeshafer
Posts: 49
Joined: Fri Feb 24, 2012 10:45 am

Angular velocity integration

Post by mikeshafer »

Hi folks,

Sorry for the noob question, but I got my boxes stacking by dividing the angular velocity by 1/2. Erin's Iterative dynamics states that the *quaternion* integration is achieved by:

q2 = q1 + dt/2 * q1 * w2

The factor of 1/2 in there seems to be required for the quaternion integration. I don't do my integration via quaternions, I do it with matrix math. Do I still need to divide by 2 for a matrix integration?

Thanks in advance,
Mike
mikeshafer
Posts: 49
Joined: Fri Feb 24, 2012 10:45 am

Re: Angular velocity integration

Post by mikeshafer »

Here is my success story*. It's got an asterisk because it's dependent on dividing angular velocity by 2, lol.

Image

Mike
mikeshafer
Posts: 49
Joined: Fri Feb 24, 2012 10:45 am

Re: Angular velocity integration

Post by mikeshafer »

Nevermind! It works even with angular velocity not divided by 2. I don't know what I did, but it does the teeter-totter that Erin describes in his paper and then stabilizes due to the caching of lambda values. Oh, and the bias calculation for the normal constraint: I'm not dividing by anything--it's straight up difference in velocity + penetration / dt. It's working and I'm happy. I can get some sleep now.

Mike
Post Reply