returns a pylist of contact points. Each point has the following data:
0 int m_contactFlags;//unused for now
1 int m_bodyUniqueIdA;
2 int m_bodyUniqueIdB;
3 int m_linkIndexA;
4 int m_linkIndexB;
5-6-7 double m_positionOnAInWS[3];//contact point location on object A, in world space coordinates
8-9-10 double m_positionOnBInWS[3];//contact point location on object A, in world space coordinates
11-12-13 double m_contactNormalOnBInWS[3];//the separating contact normal, pointing from object B towards object A
14 double m_contactDistance;//negative number is penetration, positive is distance.
15 double m_normalForce;
Small change that takes the world gravity and applies it to the
body during the call to inverse dynamics in the shared memory
interface. Otherwise the gravity vector is left at the default
value and there is currently no interface to set the gravity
for the inverse dynamics system.
Fix issue in contact/friction between btMultibody and btRigidBody (external force/torque of btRigidBody was not taken into account during contact/friction setup)
Allow 0.1 mm slop in contact, to avoid loosing contact. Todo: allow contacts with positive distance in multibody solver.
The code essentially dead because it was checking the pointer to an array for null. The check will alway return true. Instead the code was meant to check if the string had anything in it after the call to clGetDeviceInfo.
The absolute value of m_appliedImpulse was supposed to be compared to the breakingThreshold. Instead a boolean was created from the comparison, and the absolute value of that was taken (which makes no sense).
causing serialization issue in URDF/SDF loading in double-precision mode on Windows 32bit.
(it was fine on Mac/Linux 64bit)
bump serialization version to 2.85, as we change the file format
fix in name clash in makesdna.cpp -> intptr_t is already defined.
This change adds support for calculating Jacobians and
dot(Jacobian)*u terms, along with the required support for
the 3xN matrices in the standalone math library.
It also adds functions to compute kinematics only (position, velocity, accel).
To facilitate tests, the Cl also adds a RandomTreeCreator to create
randomized multibody trees.
Thanks to Thomas Buschmann for this contribution!
