Joint constraint specs

Post Reply
benelot
Posts: 350
Joined: Sat Jul 04, 2015 10:33 am
Location: Bern, Switzerland
Contact:

Joint constraint specs

Post by benelot »

I still seem to have problems when setting up different joints in bullet, so I need some specifications about the different joint constraints. Here is my question: what are the angular ranges of the different joints? Are they from -Pi to +Pi? 0 to 2Pi? I need this information to set up proper limits. Can you tell me about the following constraints?
-btHingeConstraint
-btPointToPointConstraint
-btGeneric6DofConstraint (are the spring versions the same?)
-btConeTwistConstraint
-btMultibody revolute
-btMultibody spherical (can I limit the spherical joint in a way? I feel it is not working)

Thanks for your answers!
Basroil
Posts: 463
Joined: Fri Nov 30, 2012 4:50 am

Re: Joint constraint specs

Post by Basroil »

http://ode-wiki.org/wiki/index.php?titl ... _Functions is a good place to start

For the purposes of constraint types,
odehinge = bthinge~=mb hinge
btpoint2point~=mb ball~=bt6dof==ode ball and socket
benelot
Posts: 350
Joined: Sat Jul 04, 2015 10:33 am
Location: Bern, Switzerland
Contact:

Re: Joint constraint specs

Post by benelot »

Oh ok! Thanks for your help. My setups are now working properly. I also found out that it is sometimes good for the stability of a joint to add -EPS, EPS as the joint limits when I basically need a zero limit (EPS: epsilon being a very small non-zero value).
Basroil
Posts: 463
Joined: Fri Nov 30, 2012 4:50 am

Re: Joint constraint specs

Post by Basroil »

benelot wrote:Oh ok! Thanks for your help. My setups are now working properly. I also found out that it is sometimes good for the stability of a joint to add -EPS, EPS as the joint limits when I basically need a zero limit (EPS: epsilon being a very small non-zero value).
That's usually not the best idea, since you're just hiding the problems rather than solving them. Using a different solver usually helps, as does increasing the iteration count for the sequential impulse solver. If your instability is due to high mass ratios or fast motion, double precision and higher framerates help too.
Post Reply