chenlin/Mujoco_WASM
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

Update changelog.rst for version 2.3.3

Browse Source 
PiperOrigin-RevId: 517932202
Change-Id: I55063e9bc5da2d424b1ad303834d8f11cb0a0fa9
This commit is contained in:
Nimrod Gileadi
2023-03-20 04:39:25 -07:00
committed by Copybara-Service
parent 33a2ce73d4
commit e58d53e747
1 changed files with 50 additions and 50 deletions
Download Patch File Download Diff File Expand all files Collapse all files
doc/changelog.rst
+50 -50
View File
@@ -2,83 +2,83 @@
Changelog Changelog
========= =========
Upcoming version (not yet released) Version 2.3.3 (March 20, 2023)
----------------------------------- ------------------------------
General General
^^^^^^^ ^^^^^^^
- Improvements to implicit integration: 1. Improvements to implicit integration:
- The derivatives of the RNE algorithm are now computed using sparse math, leading to significant speed - The derivatives of the RNE algorithm are now computed using sparse math, leading to significant speed
improvements for large models when using the :ref:`implicit integrator<geIntegration>`. improvements for large models when using the :ref:`implicit integrator<geIntegration>`.
- A new integrator called ``implicitfast`` was added. It is similar to the existing implicit integrator, but skips the - A new integrator called ``implicitfast`` was added. It is similar to the existing implicit integrator, but skips the
derivatives of Coriolis and centripetal forces. See the :ref:`numerical integration<geIntegration>` section for a derivatives of Coriolis and centripetal forces. See the :ref:`numerical integration<geIntegration>` section for a
detailed motivation and discussion. The implicitfast integrator is recommended for all new models and will detailed motivation and discussion. The implicitfast integrator is recommended for all new models and will
become the default integrator in a future version. become the default integrator in a future version.
The table below shows the compute cost of the 627-DoF `humanoid100 The table below shows the compute cost of the 627-DoF `humanoid100
<https://github.com/deepmind/mujoco/blob/main/model/humanoid100/humanoid100.xml>`_ model using different integrators. <https://github.com/deepmind/mujoco/blob/main/model/humanoid100/humanoid100.xml>`_ model using different integrators.
"implicit (old)" uses dense RNE derivatives, "implicit (new)" is after the sparsification mentioned above. "implicit (old)" uses dense RNE derivatives, "implicit (new)" is after the sparsification mentioned above.
Timings were measured on a single core of an AMD 3995WX CPU. Timings were measured on a single core of an AMD 3995WX CPU.
.. csv-table:: .. csv-table::
:header: "timing", "Euler", "implicitfast", "implicit (new)", "implicit (old)" :header: "timing", "Euler", "implicitfast", "implicit (new)", "implicit (old)"
:widths: auto :widths: auto
:align: left :align: left
one step (ms), 0.5, 0.53, 0.77, 5.0 one step (ms), 0.5, 0.53, 0.77, 5.0
steps/second, 2000, 1900, 1300, 200 steps/second, 2000, 1900, 1300, 200
.. image:: images/computation/midphase.gif .. image:: images/computation/midphase.gif
:align: right :align: right
:width: 350px :width: 350px
- Added a collision mid-phase for pruning geoms in body pairs, see :ref:`documentation<coSelection>` for more details. 2. Added a collision mid-phase for pruning geoms in body pairs, see :ref:`documentation<coSelection>` for more details.
This is based on static AABB bounding volume hierarchy (a BVH binary tree) in the body inertial frame. The GIF on the This is based on static AABB bounding volume hierarchy (a BVH binary tree) in the body inertial frame. The GIF on
right is cut from `this longer video <https://youtu.be/e0babIM8hBo>`_. the right is cut from `this longer video <https://youtu.be/e0babIM8hBo>`_.
- The ``mjd_transitionFD`` function no longer triggers sensor calculation unless explicitly requested. #. The ``mjd_transitionFD`` function no longer triggers sensor calculation unless explicitly requested.
- Corrected the spelling of the ``inteval`` attribute to ``interval`` in the :ref:`mjLROpt` struct. #. Corrected the spelling of the ``inteval`` attribute to ``interval`` in the :ref:`mjLROpt` struct.
- Mesh texture and normal mappings are now 3-per-triangle rather than 1-per-vertex. Mesh vertices are no longer #. Mesh texture and normal mappings are now 3-per-triangle rather than 1-per-vertex. Mesh vertices are no longer
duplicated in order to circumvent this limitation as they previously were. duplicated in order to circumvent this limitation as they previously were.
- The non-zeros for the sparse constraint Jacobian matrix are now precounted and used for matrix memory allocation. #. The non-zeros for the sparse constraint Jacobian matrix are now precounted and used for matrix memory allocation.
For instance, the constraint Jacobian matrix from the `humanoid100 For instance, the constraint Jacobian matrix from the `humanoid100
<https://github.com/deepmind/mujoco/blob/main/model/humanoid100/humanoid100.xml>`_ model, which previously required <https://github.com/deepmind/mujoco/blob/main/model/humanoid100/humanoid100.xml>`_ model, which previously required
~500,000 ``mjtNum``'s, now only requires ~6000. Very large models can now load and run with the CG solver. ~500,000 ``mjtNum``'s, now only requires ~6000. Very large models can now load and run with the CG solver.
- Modified :ref:`mju_error` and :ref:`mju_warning` to be variadic functions (support for printf-like arguments). The #. Modified :ref:`mju_error` and :ref:`mju_warning` to be variadic functions (support for printf-like arguments). The
functions :ref:`mju_error_i`, :ref:`mju_error_s`, :ref:`mju_warning_i`, and :ref:`mju_warning_s` are now deprecated. functions :ref:`mju_error_i`, :ref:`mju_error_s`, :ref:`mju_warning_i`, and :ref:`mju_warning_s` are now deprecated.
- Implemented a performant ``mju_sqrMatTDSparse`` function that doesn't require dense memory allocation. #. Implemented a performant ``mju_sqrMatTDSparse`` function that doesn't require dense memory allocation.
- Added ``mj_stackAllocInt`` to get correct size for allocating ints on mjData stack. Reducing stack memory usage #. Added ``mj_stackAllocInt`` to get correct size for allocating ints on mjData stack. Reducing stack memory usage
by 10% - 15%. by 10% - 15%.
Python bindings Python bindings
^^^^^^^^^^^^^^^ ^^^^^^^^^^^^^^^
- Fixed IPython history corruption when using ``viewer.launch_repl``. The ``launch_repl`` function now provides seamless 10. Fixed IPython history corruption when using ``viewer.launch_repl``. The ``launch_repl`` function now provides
continuation of an IPython interactive shell session, and is no longer considered experimental feature. seamless continuation of an IPython interactive shell session, and is no longer considered experimental feature.
- Added ``viewer.launch_passive`` which launches the interactive viewer in a passive, non-blocking mode. Calls to #. Added ``viewer.launch_passive`` which launches the interactive viewer in a passive, non-blocking mode. Calls to
``launch_passive`` return immediately, allowing user code to continue execution, with the viewer automatically ``launch_passive`` return immediately, allowing user code to continue execution, with the viewer automatically
reflecting any changes to the physics state. (Note that this functionality is currently in experimental/beta stage, reflecting any changes to the physics state. (Note that this functionality is currently in experimental/beta stage,
and is not yet described in our :ref:`viewer documentation<PyViewer>`.) and is not yet described in our :ref:`viewer documentation<PyViewer>`.)
- Added the ``mjpython`` launcher for macOS, which is required for ``viewer.launch_passive`` to function there. #. Added the ``mjpython`` launcher for macOS, which is required for ``viewer.launch_passive`` to function there.
- Removed ``efc_`` fields from joint indexers. Since the introduction of arena memory, these fields now have dynamic #. Removed ``efc_`` fields from joint indexers. Since the introduction of arena memory, these fields now have dynamic
sizes that change between time steps depending on the number of active constraints, breaking strict correspondence sizes that change between time steps depending on the number of active constraints, breaking strict correspondence
between joints and ``efc_`` rows. between joints and ``efc_`` rows.
- Added a number of missing fields to the bindings of ``mjVisual`` and ``mjvPerturb`` structs. #. Added a number of missing fields to the bindings of ``mjVisual`` and ``mjvPerturb`` structs.
Simulate Simulate
^^^^^^^^ ^^^^^^^^
- Implemented a workaround for `broken VSync <https://github.com/glfw/glfw/issues/2249>`_ on macOS so that the frame 15. Implemented a workaround for `broken VSync <https://github.com/glfw/glfw/issues/2249>`_ on macOS so that the frame
rate is correctly capped when the Vertical Sync toggle is enabled. rate is correctly capped when the Vertical Sync toggle is enabled.
.. image:: images/changelog/contactlabel.png .. image:: images/changelog/contactlabel.png
:align: right :align: right
:width: 400px :width: 400px
- Added optional labels to contact visualization, indicating which two geoms are contacting (names if defined, ids 16. Added optional labels to contact visualization, indicating which two geoms are contacting (names if defined, ids
otherwise). This can be useful in cluttered scenes. otherwise). This can be useful in cluttered scenes.
|br| |br|