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

Remove native sensor noise sampling.

Browse Source 
PiperOrigin-RevId: 620075316
Change-Id: I6a0a1d63834e7c8bfd290f89d86fea791d794739
This commit is contained in:
Yuval Tassa
2024-03-28 15:31:54 -07:00
committed by Copybara-Service
parent 47ba72ea59
commit 5d26b50fce
14 changed files with 54 additions and 135 deletions
Download Patch File Download Diff File Expand all files Collapse all files
doc/changelog.rst
+26 -10
View File
@@ -7,29 +7,45 @@ Upcoming version (not yet released)
General
^^^^^^^
1. Added the :ref:`actuatorgravcomp<body-joint-actuatorgravcomp>` joint attribute. When enabled, gravity compensation
.. admonition:: Breaking API changes
:class: attention
1. Removed the ability to natively add noise to sensors. Note that the ``mjModel.sensor_noise`` field and
:ref:`corresponding attribute<CSensor>` are kept and now function as a convenient location for the user to save
standard-deviation information for their own use. This feature was removed because:
- There was no mechanism to seed the randon noise generator.
- It was not thread-safe, even if seeding would have been provided, sampling on multiple threads would lead to
non-reproducible results.
- This feature was seen as overreach by the engine. Adding noise should be the user's responsibility.
- We are not aware of anyone who was actually using the feature.
**Migration:** Add noise to sensor values yourself.
2. Added the :ref:`actuatorgravcomp<body-joint-actuatorgravcomp>` joint attribute. When enabled, gravity compensation
forces on the joint are treated as applied by actuators. See attribute documentation for more details. The example
model
`refsite.xml <https://github.com/google-deepmind/mujoco/blob/main/test/engine/testdata/actuation/refsite.xml>`__,
which demostrates Cartesian actuation of an arm, has been updated to use this attribute.
2. Added support for gmsh format 2.2, as generated by e.g. `fTetwild <https://github.com/wildmeshing/fTetWild>`__.
3. Added support for gmsh format 2.2, as generated by e.g. `fTetwild <https://github.com/wildmeshing/fTetWild>`__.
MJX
^^^
3. Improved performance of SAT for convex collisions.
4. Fixed bug for sphere/capsule-convex deep penetration.
5. Fixed bug where ``mjx.Data`` produced by ``mjx.put_data`` had different treedef than ``mjx.make_data``.
6. Throw an error for margin/gap for convex mesh collisions, since they are not supported.
4. Improved performance of SAT for convex collisions.
5. Fixed bug for sphere/capsule-convex deep penetration.
6. Fixed bug where ``mjx.Data`` produced by ``mjx.put_data`` had different treedef than ``mjx.make_data``.
7. Throw an error for margin/gap for convex mesh collisions, since they are not supported.
Simulate
^^^^^^^^
7. Fixed bug in order of enable flag strings. Before this change, using the simulate UI to toggle the
:ref:`invdiscrete<option-flag-invdiscrete>` or :ref:`sensornoise<option-flag-sensornoise>`
flags would actually toggle the other flag.
8. Fixed bug in order of enable flag strings. Before this change, using the simulate UI to toggle the
:ref:`invdiscrete<option-flag-invdiscrete>` or the (now removed) ``sensornoise`` flags would actually toggle the
other flag.
Python bindings
^^^^^^^^^^^^^^^
8. Added the ``mujoco.minimize`` Python module for nonlinear least-squares, designed for System Identification (sysID).
9. Added the ``mujoco.minimize`` Python module for nonlinear least-squares, designed for System Identification (sysID).
The sysID tutorial is work in progress, but a pedagogical colab notebook with examples, including Inverse
Kinematics, is available here: |ls_colab|