[ODE] Jittering of light weighted model

[Jon Watte (ODE)]

Daniel Hein wrote:

> After modeling everything in a meter-kg-second system I encountered 
> problems with jittering of the whole robot when it just stands on the 
> ground. As I understand this is caused by the collision handling. I 

 > 1. How can I adjust ODE to the (real) weighted robot model, in that way
 > that there is no visible jittering - without touching the dWorldStep
 > time? (or in other words: How can I adjust ODE that the absolute
 > amplitude of jittering of the light weighted model is the same as with
 > the 10times weighted robot?)

90 times of 100 this is because you're not using a fixed time step. Make 
sure you're using a fixed time step a la 
http://www.mindcontrol.org/~hplus/graphics/game_loop.html before you try 
for any other solution. (It sounds like you might already, though)

9 times out of 100, jitter is caused by not applying rotational inertia 
dampening on your bodies. Each time step, get the rotational velocity, 
and add back -0.01 times that velocity as counter-inertia. This will do 
wonders for stabilizing simulations, without noticeably affecting the 
visible fidelity of the simulation negatively.

If you have the 1 time of 100 that's harder, then I would look at the 
collision joints -- are you creating enough joints for each body? Are 
you coalescing coincident contacts? Do your friction forces look 
reasonable? How about joint constraints? A little less friction will 
often improve stability, because it allows feet to slide when there are 
kinematic loops between the body and the ground.

> the 10times (so using a meter-100g-second system): Except adjusting the 
> forces/torques of all joints, what else do I had to attend to? If in 

That's pretty much it. You might want to change any dampening/friction 
forces you have, to compensate for the larger inertia.

> real life I would scale the robots mass to the 10times, the whole body 
> caracteristics (e.g. body frequency) would change - so if I just scale 

ODE does not simulate internal resonance of the bodies; they are assumed 
to be perfectly rigid, so the mass does not affect any "material" 
attributes.

Cheers,

			/ h+