update to OpenVR sdk 1.03 from https://github.com/ValveSoftware/openvr
add camPosX/Y/Z and camRotZ to adjust relative camera/world transform for VR (so you can align virtual table with real table etc)
tweak quadruped.py to move a bit
add mouse picking to physics server
add UDP network connection for physics client <-> server.
also set spinning friction in rolling friction demo (otherwise objects may keep on spinning forever)
URDF/SDF: add a flag to force concave mesh collisiofor static objects. <collision concave="yes" name="pod_collision">
VR: support teleporting using buttong, allow multiple controllers to be used, fast wireframe rendering,
Turn off warnings about deprecated C routine in btScalar.h/b3Scalar.h
Add a dummy return to stop a warning
Expose defaultContactERP in shared memory api/pybullet.
First start to expose IK in shared memory api/pybullet (not working yet)
Expose inverse dynamics to Bullet shared memory API, through b3CalculateInverseDynamicsCommandInit and
b3GetStatusInverseDynamicsJointForces command/status. See PhysicsClientExeample or pybullet for usage.
Add option for Windows and Linux to set python_lib_dir and python_include_dir for premake and --enable_pybullet option
Expose inverse dynamics to pybullet: [force] = p.calculateInverseDynamics(objectIndex,[q],[qdot],[acc])
Thanks to Jeff Bingham for the suggestion.
Disable a few Visual Studio warnings:
C4244: conversion from 'type1' to 'type2', possible loss of data
C4267: conversion from 'size_t' to 'type', possible loss of data
Considering the GCC/Clang counterpart '-Wconversion'
(https://gcc.gnu.org/wiki/NewWconversion) is not covered by '-Wall', it
seems to make more sense to disable these two warnings in Visual Studio,
rather than fixing these warnings with hundreds of static_cast<> in the
source code.
it works without OpenGL dependency now, so it runs in the cloud :-)
Add scaling support for TinyRenderer, remove some un-used normal mapping in TinyRenderer shader, expose light_dir_world,
remove accidental hard-coded path in tga write_tga_file,
Fix InverseDynamicsExampleCreateFunc, enum has to start at 0
(for now, start the example browser in 'physics server',
then compile using premake --python option,
then run python in the bin folder (so it finds pybullet.so)
and run the test.py script in examples/pybullet folder.
The robotics shared memory C API is very suitable for this.
remove targetdir from all libraries in premake, so it is much easier to create a separate folder for all binary+lib
transmit the serialized btMultiBody data back from server to client, after the server loads a URDF file. This includes base+link+joint names
tweak the serialization routines, so it is easier to skip pointers and to serialize directly to a shared memory buffer
also tweak the serialization code to allow to process data without 'DNA' schema data (assuming file-DNA = memory DNA)
add example description for all examples (with word-wrap)
add the VoronoiFractureDemo, note that the collision are disabled after breaking constraints.
add optional GwenOpenGLTest, to make it easier to see Gwen user interface features.
add MultiBody Custom Creation example, to show how to import data from a URDF file and fill up your own data structures.
add btMultiBody::setBaseWorldTransform method
todo: fix cmake build, this patch is premake only
add sphere2.urdf
move btSpatialAlgebra into LinearMath
remove some warnings, introduce BT_ZERO, BT_ONE, BT_HALF as defines for 0.f/0., 1.f/1., 0.5f/0.5 respectively
--mp4=video.mp4 note that you have to re-convert for Quicktime
ffmpeg -f mp4 -vcodec h264 -i test.mp4 -pix_fmt yuv420p test2.mp4
add the option to display text in 3d, used in Coriolis demo
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.
