[ODE] Bounciness

Bryan Thibodeau thibodea at cs.umass.edu
Tue Jul 11 06:27:51 MST 2006


As others have mentioned, ground penetration is a likely culprit.  
Another possibility is not so much the velocity being generated by the 
amotors, but the force being applied at the joints to achieve that 
velocity.  If I recall correctly, an amotor will apply as much force as 
necessary (up to the limit set FMax for that joint, which I think 
defaults to dInfinity) to a joint to achieve the desired velocity in one 
time step.  If you have a very small step size, as you likely would for 
simulating something like bipedal walking, there may be a very large 
impulse force generated initially by the amotor, causing the robot to 
hop.  You could try setting the FMax parameter of the amotor to a 
reasonable value, where reasonable will depend heavily on the type of 
motor and gear box that you assume and the actual velocity of the motor.

Bryan


Bob Prog wrote:
> Hello,
>
> I am trying to simulate a walking biped using ODE.  When I set my amotor 
> velocities to try and take a step, the biped immediately hops into the air 
> in a forward motion.  I don't think the velocities I'm applying are so great 
> that it is warranting such a hop.  I would like to just cruise along the 
> ground like a normal walk.  I am not setting the dContactBounce flag in the 
> collision callback, and I'm not using springs.  So what could be causing 
> this?  What am I missing here?  Or is it probably the velocities?
>
> Thanks,
> Bob
>
> _________________________________________________________________
> Express yourself instantly with MSN Messenger! Download today - it's FREE! 
> http://messenger.msn.click-url.com/go/onm00200471ave/direct/01/
>
> _______________________________________________
> ODE mailing list
> ODE at q12.org
> http://q12.org/mailman/listinfo/ode
>   



More information about the ODE mailing list