Option to de/serialize btPersistentContactManifolds and fix lossy conversion during btMultiBody de/serialization of base world transform
(serialize the quaternion, not the converted 3x3 matrix)
There are still several caches not taken into account, and btMultiBody links/constraints are not deserialized yet etc.
See examples\pybullet\examples\saveRestoreState.py for an example.
implement friction anchors, position friction correction, disabled by default. Use colObj->setCollisionFlag(flag | CF_HAS_FRICTION_ANCHOR); See test/RobotClientAPI/SlopeFrictionMain.cpp. In URDF or SDF, add <friction_anchor/> in <contact> section of <link> to enable.
PhysicsServer: properly restore old activation state after releasing picked object
btMultiBodyConstraintSolver: disable flip/flop of contact/friction constraint solving by default (it breaks some internal flaky unit tests)
fixed some more warnings
added alignment macros to some classes
btPersistentManifold from 128 to 16 bytes aligned
prepare command to select collision filter mode (SIM_PARAM_UPDATE_COLLISION_FILTER_MODE)
Added demos for rigid and multi body soft (compliant) contact.
Will also add simplified Hertz compliant contact, by dynamically modifying the ERP/CFM to mimic a non-linear spring.
Note that btManifoldPoint is growing too big, we need to implement proper contact constraints derived from btTypedConstraint.
CFM for contacts use world CFM value by default, and can override with custom CFM value using the
BT_CONTACT_FLAG_HAS_CONTACT_CFM stored in m_contactPointFlags.
Boolean m_lateralFrictionInitialized is replaced 'BT_CONTACT_FLAG_LATERAL_FRICTION_INITIALIZED' flag stored in int m_contactPointFlags in btManifoldPoint.
Enable successive over-relaxation parameter (SOR) for contacts. btMLCPSolver uses global CFM.
In one of the next commits, contact softness will be enabled btMultiBody contacts.
Also need to review use of CFM in btMLCPSolvers (only world CFM is used at the moment)
workaround for reversed separating normal in gjk/epa when using very small shapes, detect case and revert normal.
use smaller world size (10 units versus 100) for higher resolution shadow map
use a hard-coded rolling friction of 0.2 for objects in physics server (will make this configurable)
fix loading of command log files, when platform features are different (64bit/32bit)
to the rest of the Bullet code base. This code will replace the original GJK/EPA
in a future commit.
Added btMprPenetration, an implementation of Minkowski Portal Refinement
by Daniel Fiser. Original MPR idea is by Gary Snethen, and the first
implementation is here: https://github.com/erwincoumans/xenocollide
It is an alternative to EPA, although computing the local penetration depth.
EPA computes the global penetration depth. In many cases, MPR is sufficient
and performs better than EPA.
Add btCollisionObject::setIgnoreCollisionCheck to disable collisions between specific instances, without having a btTypedConstraint. See https://github.com/bulletphysics/bullet3/issues/165
Make btMultiBody and btMultiBodyJointMotor backwards compatible with Bullet 2.82 API (single-DOF API)
rayResult.m_flags |= btTriangleRaycastCallback::kF_UseGjkConvexRaytest;
See Demos/RaytestDemo/RaytestDemo.cpp for usage
This fixes issue 146, longest outstanding from 2008.
In some degenerate cases the contact normal is pointing the wrong direction
so fix it now (until we can deal with all degenerate cases in GJK and EPA)
contact normals need to point from B to A in all cases, so we can simply check if the contact normal really points from B to A
We like to use a dot product of the normal against the difference of the centroids,
once the centroid is available in the API
until then we use the center of the aabb to approximate the centroid
add compound shape support to BulletXmlWorldImporter and fix some compile issue under Debian (hopefully)
object picking change in the demos: create a ball-socket picking constraint when holding shift while mouse dragging, otherwise a fixed (6dof) constraint
add assert in constraint solver, when both objects have their inertia tensor rows set to zero
btPolyhedralContactClipping: add edge-edge contact point in findSeparatingAxis (similar to the default GJK case)
See http://www.youtube.com/watch?v=RV7sBAsKu4M and Bullet/Demos/RollingFrictionDemo
Fixes in FractureDemo (mouse picking constraint needs to be removed, otherwise constraint solver crashes/asserts)
This should improve 'ccd' handling when using world->getDispatchInfo().m_useContinuous = true;body->setCcdSquareMotionThreshold(...); body->setCcdSquareMotionThreshold(...)
shoot smaller boxes (test)
use yellow instead of orange for contact point normals
tweak default erp and erp2 values, now split impulse is on by default (need to check it)
warmstarting for contact points was broken, fix in btPersistentManifold
enable split impulse by default (at the cost of some performance)
add the option for zero-length friction (instead of recomputing friction directions using btPlaneSpace), use the solver mode flag SOLVER_ALLOW_ZERO_LENGTH_FRICTION_DIRECTIONS
precompute lateral friction directions (in btManifoldResult)
remove the mConstraintRow[3] from btManifoldPoint, it just took a lot of memory with no benefits: fixed it in btParallelConstraintSolver
For OSX:
cd build
./premake_osx xcode4
for iOS:
cd build
./ios_build.sh
./ios_run.sh
Also integrated the branches/StackAllocation to make it easier to multi-thread collision detection in the near future. It avoids changing the btCollisionObject while performing collision detection.
As this is a large patch, some stuff might be temporarily broken, I'll keep an eye out on issues.
