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

Add LQR tutorial colab notebook.

Browse Source 
PiperOrigin-RevId: 475551810
Change-Id: Iac8ca4fff88331f325e34fef01b41307603a6416
This commit is contained in:
Yuval Tassa
2022-09-20 07:40:30 -07:00
committed by Copybara-Service
parent f13a58326d
commit f151e84a1d
2 changed files with 1126 additions and 6 deletions
Download Patch File Download Diff File Expand all files Collapse all files
doc/changelog.rst
+16 -6
View File
@@ -9,12 +9,20 @@ Upcoming version (not yet released)
General General
^^^^^^^ ^^^^^^^
.. youtube:: RHnXD6uO3Mg
:align: right
:height: 150px
- Added colab notebook tutorial showing how to balance the humanoid on one leg with a Linear Quadratic Regulator. The
notebook uses MuJoCo's native Python bindings, and includes a draft ``Renderer`` class, for easy rendering in Python.
|br| Try it yourself: |LQRopenincolab|
- Updates to humanoid model: - Updates to humanoid model:
- Added two keyframes (stand-on-one-leg and squat). - Added two keyframes (stand-on-one-leg and squat).
- Increased maximum hip flexion angle. - Increased maximum hip flexion angle.
- Added hamstring tendons which couple the hip and knee at high hip flexion angles. - Added hamstring tendons which couple the hip and knee at high hip flexion angles.
- General cosmetic improvements, including improved use of defaults and better naming scheme. - General cosmetic improvements, including improved use of defaults and better naming scheme.
- Added :ref:`mju_boxQP` and allocation function :ref:`mju_boxQPmalloc` for solving the box-constrained - Added :ref:`mju_boxQP` and allocation function :ref:`mju_boxQPmalloc` for solving the box-constrained
Quadratic Program: Quadratic Program:
@@ -23,8 +31,8 @@ General
x^* = \text{argmin} \; \tfrac{1}{2} x^T H x + x^T g \quad \text{s.t.} \quad l \le x \le u x^* = \text{argmin} \; \tfrac{1}{2} x^T H x + x^T g \quad \text{s.t.} \quad l \le x \le u
- The algorithm, introduced in `Tassa et al. 2014 <https://doi.org/10.1109/ICRA.2014.6907001>`_, The algorithm, introduced in `Tassa et al. 2014 <https://doi.org/10.1109/ICRA.2014.6907001>`_,
converges after 2-5 Cholesky factorisations, independent of problem size. converges after 2-5 Cholesky factorisations, independent of problem size.
- Added :ref:`mju_mulVecMatVec` to multiply a square matrix :math:`M` with vectors :math:`x` and :math:`y` on both - Added :ref:`mju_mulVecMatVec` to multiply a square matrix :math:`M` with vectors :math:`x` and :math:`y` on both
sides. The function returns :math:`x^TMy`. sides. The function returns :math:`x^TMy`.
@@ -35,6 +43,8 @@ General
as global overrides. Note the new mechanism is currently undocumented except in code, as we test it internally. as global overrides. Note the new mechanism is currently undocumented except in code, as we test it internally.
If you are interested in using the pluging mechanism, please get in touch first. If you are interested in using the pluging mechanism, please get in touch first.
.. |LQRopenincolab| image:: https://colab.research.google.com/assets/colab-badge.svg
:target: https://colab.research.google.com/github/deepmind/mujoco/blob/main/python/LQR.ipynb
Version 2.2.2 (September 7, 2022) Version 2.2.2 (September 7, 2022)
--------------------------------- ---------------------------------
python/mujoco/LQR.ipynb
+1110
View File
File diff suppressed because it is too large Load Diff