Hill-Type Muscle - Non linear spring
-
- Posts: 7
- Joined: Tue Jan 19, 2016 1:51 pm
Hill-Type Muscle - Non linear spring
Hi all, I am attempting to create hill type models for my computing project but I can't quite work out how to create a non linear spring in bullet and am hoping someone has some advice
-
- Posts: 13
- Joined: Sun Oct 25, 2015 10:17 am
Re: Hill-Type Muscle - Non linear spring
I'm working on a similar thing. My current attempt is briefly described here, posted a few days ago: http://bulletphysics.org/Bullet/phpBB3/ ... =9&t=11022
It is not exactly the same as the hill-type muscle, but a variation described in this paper: http://www.livingcontrolsystems.com/int ... _model.pdf
There are three small spherical bodies, A, B and C. A and C are connected to 'bones' with point constraints. A and B are connected with a btGeneric6dofSpring2Constraint modeling the tendon, while B and C are connected with a slider constraint. I've tried setting the slider motor target velocity, and that sort-of works. To make the model more non-linear, I've turned off the motor and I'm trying to apply forces directly to the B connector to move toward the C body along the slider constraint in order to model the nonlinear damping of the muscle. I have a feeling this could be accomplished in different ways, maybe with just two boides and a single slider constraint, but I'll leave that for later.
It is not exactly the same as the hill-type muscle, but a variation described in this paper: http://www.livingcontrolsystems.com/int ... _model.pdf
There are three small spherical bodies, A, B and C. A and C are connected to 'bones' with point constraints. A and B are connected with a btGeneric6dofSpring2Constraint modeling the tendon, while B and C are connected with a slider constraint. I've tried setting the slider motor target velocity, and that sort-of works. To make the model more non-linear, I've turned off the motor and I'm trying to apply forces directly to the B connector to move toward the C body along the slider constraint in order to model the nonlinear damping of the muscle. I have a feeling this could be accomplished in different ways, maybe with just two boides and a single slider constraint, but I'll leave that for later.