[ODE] movement force conversion

Piotr Obrzut piotr_obrzut at o2.pl
Tue Mar 15 10:11:56 MST 2005


Hi,

11 marca 2005 (22:08:12) Jon wrote:


>> thanks, but the problem is that i don't use ODE while this movement -
>> animations controls my model then. For instance: after calling "kick"
>> animation and positive collision checking I wan't to add apropriate
>> force into hited body.

JW> The ODE collision system generates forces based on relative 
JW> velocities, bounciness, and penetration depth. Basically, the 
JW> idea is to add opposing forces that drive the objects out of 
JW> each other in one time step.

AM> Did you try to, at the time of impact in the animation, compute the
AM> instantaneous velocity of the fist/foot by measuring the distance it has
AM> moved in the animation since the last time step?  Then you would just
AM> multiply the mass of the fist/foot with the velocity and get the impact
AM> momentum.

AM> --Adam

I was thinking that when move is more complex (due to involved human
body parts when kicking) then this problem will be also complex.
But after reading mails from Jon Watte and Adam Moravanszky I've
noticed that this problem reduce only to 2 collided bodies and their after
collision behavior and I can forget about other bodies, am I right?

JW> However, you can use ODE with animation. Just make the animations 
JW> turn into forward kinematic forces, rather than absolute joint 
JW> position/orientations. This makes the right thing happen if you 
JW> run into a joint, etc.

forward kinematic forces? do you mean forward dynamic forces?

that is what interest me in longer time, but now before any
research: it is possible to turn stored forward kinemtic animation
into forward dynamics? I was always thinking that kinemtic move is a
result of dynamics so it could be realy hard to make sth backward.
Simply said I want to do know whether it could bo possible to convert
animations like in Cal3d into forward dynamic animation. Sorry If this
sounds stupid, but I didn't read any precise docs about forward
dynamics at all but I intend to do this in the nearest future.

-- 
greetings,
 Piotr Obrzut



More information about the ODE mailing list