[ODE] Car Braking Problem

[Jon Watte (ODE)]

If you're not setting the fdir1 correctly (and setting the flag in the
contact) then that's almost surely your problem. Carworld, the ODE car
sample application with source, does this; you could grep the source:

http://www.mindcontrol.org/~hplus/carworld/

Cheers,

			/ h+


KuRi wrote:
> Thanks for your answers. It looks (like you say) like a contact problem,
> because with no contacts the problem dissapear. But i don't understand why
> the error depends on the Y Angle of the car. Perhaps it is a round problem.
> Can someone post one example with code of setting the fdir1 to the rolling
> tyre direction, etc...? I have tried that with no results...
> 
> Thanks again. KuRiX
> 
> ----- Original Message ----- 
> From: "Matthew D. Hancher" <mdh at email.arc.nasa.gov>
> To: <ode at q12.org>
> Sent: Monday, June 27, 2005 10:23 PM
> Subject: [ODE] Re: Car Braking Problem
> 
> 
> 
>>>When i brake (dBodyAddTorque), the car turns to the right or to the
>>>left
>>>depending on the Y Axis Angle.
>>
>>Perhaps this is a product of ODE's friction model, which is simplified
>>in ways that make it anisotropic.  You can address this somewhat by
>>cleverly setting the contact's friction direction, rather than letting
>>ODE pick it arbitrarily.  When you generate each contact joint between
>>a wheel and the ground, set the fdir1 field of the dContact structure to
>>be a unit vector that is in the direction of horizontal motion of the
>>wheel geom, and set the dContactFDir1 bit in the contact flags.
>>
>>Matt
>>
>>Matt Hancher
>>Intelligent Systems Division
>>NASA Ames Research Center
>>mdh at email.arc.nasa.gov
>>
>>_______________________________________________
>>ODE mailing list
>>ODE at q12.org
>>http://q12.org/mailman/listinfo/ode
> 
> 
> _______________________________________________
> ODE mailing list
> ODE at q12.org
> http://q12.org/mailman/listinfo/ode
> 
>