Restructures documentation and fixes class types (#98)

This change restructures documentation to add a new Overview section
that combines previous pages around learning workflows, environment
design, and developer guides.

New pages are added for RL library comparison and training performance
benchmarking.

<!-- As you go through the list, delete the ones that are not
applicable. -->

- Bug fix (non-breaking change which fixes an issue)
- New feature (non-breaking change which adds functionality)
- This change requires a documentation update

- [x] I have run the [`pre-commit` checks](https://pre-commit.com/) with
`./isaaclab.sh --format`
- [x] I have made corresponding changes to the documentation
- [x] My changes generate no new warnings
- [ ] I have added tests that prove my fix is effective or that my
feature works
- [ ] I have updated the changelog and the corresponding version in the
extension's `config/extension.toml` file
- [x] I have added my name to the `CONTRIBUTORS.md` or my name already
exists there

<!--
As you go through the checklist above, you can mark something as done by
putting an x character in it

For example,
- [x] I have done this task
- [ ] I have not done this task
-->

docs/index.rst

@@ -6,9 +6,19 @@ Overview
    :alt: H1 Humanoid example using Isaac Lab
 
 **Isaac Lab** is a unified and modular framework for robot learning that aims to simplify common workflows
-in robotics research (such as RL, learning from demonstrations, and motion planning). It is built upon
+in robotics research (such as reinforcement learning, learning from demonstrations, and motion planning). It is built upon
 `NVIDIA Isaac Sim`_ to leverage the latest simulation capabilities for photo-realistic scenes, and fast
-and efficient simulation. The core objectives of the framework are:
+and efficient simulation.
+
+Isaac Lab provides over 26 environments, and we are actively working on adding more environments to the list. Current environments include Classic tasks - Cartpole, Cartpole with camera, Humanoid and Ant, fixed-arm and dexterous manipulation tasks - UR10, Franka, Allegro, Shadow Hand, Legged locomotion tasks - Quadrupeds and Humanoids, Navigation tasks - Quadruped, and Quadcopter. It features over 16 robots:
+
+- Classic - Cartpole, Humanoid, Ant
+- Fixed-Arms and Hands - UR10, Franka, Allegro, Shadow Hand
+- Quadrupeds – Anybotics Anymal-B, Anymal-C, Anymal-D, Unitree A1, Unitree Go1, Unitree Go2, Boston Dynamics Spot
+- Humanoids - Unitree H1, Unitree G1
+- Quadcopter - Crazyflie
+
+The core objectives of the framework are:
 
 - **Modularity**: Easily customize and add new environments, robots, and sensors.
 - **Agility**: Adapt to the changing needs of the community.
@@ -60,23 +70,18 @@ Table of Contents
    :caption: Getting Started
 
    source/setup/installation/index
-   source/setup/developer
-   source/setup/sample
-   source/setup/template
    source/setup/faq
 
 .. toctree::
-   :maxdepth: 2
-   :caption: Features
+   :maxdepth: 3
+   :caption: Overview
+   :titlesonly:
 
-   source/features/task_workflows
-   source/features/hydra
-   source/features/multi_gpu
-   source/features/tiled_rendering
-   source/features/environments
-   source/features/actuators
-   source/features/reproducibility
-   .. source/features/motion_generators
+   source/overview/developer-guide/index
+   source/overview/environment-workflows/index
+   source/overview/reinforcement-learning/index
+   source/overview/teleop_imitation
+   source/overview/sample_scripts
 
 .. toctree::
    :maxdepth: 1
@@ -87,12 +92,6 @@ Table of Contents
    source/how-to/index
    source/deployment/index
 
-.. toctree::
-   :maxdepth: 1
-   :caption: Source API
-
-   source/api/index
-
 .. toctree::
    :maxdepth: 1
    :caption: Migration Guides
@@ -102,6 +101,25 @@ Table of Contents
    source/migration/migrating_from_omniisaacgymenvs
    source/migration/migrating_from_orbit
 
+.. toctree::
+   :maxdepth: 3
+   :caption: Features
+
+   source/features/task_workflows
+   source/features/hydra
+   source/features/multi_gpu
+   source/features/tiled_rendering
+   source/features/environments
+   source/features/actuators
+   source/features/reproducibility
+   .. source/features/motion_generators
+
+.. toctree::
+   :maxdepth: 1
+   :caption: Source API
+
+   source/api/index
+
 .. toctree::
    :maxdepth: 1
    :caption: References

docs/source/features/environments.rst

@@ -75,7 +75,8 @@ for the reach environment:
     +----------------+---------------------------+-----------------------------------------------------------------------------+
     | |lift-cube|    | |lift-cube-link|          | Pick a cube and bring it to a sampled target position with the Franka robot |
     +----------------+---------------------------+-----------------------------------------------------------------------------+
-    | |cabi-franka|  | |cabi-franka-link|        | Grasp the handle of a cabinet's drawer and open it with the Franka robot    |
+    | |cabi-franka|  | | |cabi-franka-link|      | Grasp the handle of a cabinet's drawer and open it with the Franka robot    |
+    |                | | |franka-direct-link|    |                                                                             |
     +----------------+---------------------------+-----------------------------------------------------------------------------+
     | |cube-allegro| | |cube-allegro-link|       | In-hand reorientation of a cube using Allegro hand                          |
     |                | |allegro-direct-link|     |                                                                             |
@@ -98,6 +99,7 @@ for the reach environment:
 .. |lift-cube-ik-abs-link| replace:: `Isaac-Lift-Cube-Franka-IK-Abs-v0 <https://github.com/isaac-sim/IsaacLab/blob/main/source/extensions/omni.isaac.lab_tasks/omni/isaac/lab_tasks/manager_based/manipulation/lift/config/franka/ik_abs_env_cfg.py>`__
 .. |lift-cube-ik-rel-link| replace:: `Isaac-Lift-Cube-Franka-IK-Rel-v0 <https://github.com/isaac-sim/IsaacLab/blob/main/source/extensions/omni.isaac.lab_tasks/omni/isaac/lab_tasks/manager_based/manipulation/lift/config/franka/ik_rel_env_cfg.py>`__
 .. |cabi-franka-link| replace:: `Isaac-Open-Drawer-Franka-v0 <https://github.com/isaac-sim/IsaacLab/blob/main/source/extensions/omni.isaac.lab_tasks/omni/isaac/lab_tasks/manager_based/manipulation/cabinet/config/franka/joint_pos_env_cfg.py>`__
+.. |franka-direct-link| replace:: `Isaac-Franka-Cabinet-Direct-v0 <https://github.com/isaac-sim/IsaacLab/blob/main/source/extensions/omni.isaac.lab_tasks/omni/isaac/lab_tasks/direct/franka_cabinet/franka_cabinet_env.py>`__
 .. |cube-allegro-link| replace:: `Isaac-Repose-Cube-Allegro-v0 <https://github.com/isaac-sim/IsaacLab/blob/main/source/extensions/omni.isaac.lab_tasks/omni/isaac/lab_tasks/manager_based/manipulation/inhand/config/allegro_hand/allegro_env_cfg.py>`__
 .. |allegro-direct-link| replace:: `Isaac-Repose-Cube-Allegro-Direct-v0 <https://github.com/isaac-sim/IsaacLab/blob/main/source/extensions/omni.isaac.lab_tasks/omni/isaac/lab_tasks/direct/allegro_hand/allegro_hand_env_cfg.py>`__
 

docs/source/how-to/wrap_rl_env.rst

@@ -10,7 +10,7 @@ Environment wrappers are a way to modify the behavior of an environment without
 This can be used to apply functions to modify observations or rewards, record videos, enforce time limits, etc.
 A detailed description of the API is available in the :class:`gymnasium.Wrapper` class.
 
-At present, all RL environments inheriting from the :class:`~envs.ManagerBasedRLEnv` class
+At present, all RL environments inheriting from the :class:`~envs.ManagerBasedRLEnv` or :class:`~envs.DirectRLEnv` classes
 are compatible with :class:`gymnasium.Wrapper`, since the base class implements the :class:`gymnasium.Env` interface.
 In order to wrap an environment, you need to first initialize the base environment. After that, you can
 wrap it with as many wrappers as you want by calling ``env = wrapper(env, *args, **kwargs)`` repeatedly.
@@ -127,7 +127,7 @@ Every learning framework has its own API for interacting with environments. For
 `Stable-Baselines3`_ library uses the `gym.Env <https://gymnasium.farama.org/api/env/>`_
 interface to interact with environments. However, libraries like `RL-Games`_, `RSL-RL`_ or `SKRL`_
 use their own API for interfacing with a learning environments. Since there is no one-size-fits-all
-solution, we do not base the :class:`~envs.ManagerBasedRLEnv` class on any particular learning framework's
+solution, we do not base the :class:`~envs.ManagerBasedRLEnv` and :class:`~envs.DirectRLEnv` classes on any particular learning framework's
 environment definition. Instead, we implement wrappers to make it compatible with the learning
 framework's environment definition.
 
@@ -155,6 +155,7 @@ Adding new wrappers
 
 All new wrappers should be added to the :mod:`omni.isaac.lab_tasks.utils.wrappers` module.
 They should check that the underlying environment is an instance of :class:`omni.isaac.lab.envs.ManagerBasedRLEnv`
+or :class:`~envs.DirectRLEnv`
 before applying the wrapper. This can be done by using the :func:`unwrapped` property.
 
 We include a set of wrappers in this module that can be used as a reference to implement your own wrappers.

docs/source/setup/developer.rst → docs/source/overview/developer-guide/extensions.rst

-Developer's Guide
-=================
-
-For development, we suggest using `Microsoft Visual Studio Code
-(VSCode) <https://code.visualstudio.com/>`__. This is also suggested by
-NVIDIA Omniverse and there exists tutorials on how to `debug Omniverse
-extensions <https://www.youtube.com/watch?v=Vr1bLtF1f4U&ab_channel=NVIDIAOmniverse>`__
-using VSCode.
-
-
-Setting up Visual Studio Code
------------------------------
-
-The following is only applicable for Isaac Sim installed via the Omniverse Launcher.
-The ``Isaac Lab`` repository includes the VSCode settings to easily allow setting
-up your development environment. These are included in the ``.vscode`` directory
-and include the following files:
-
-.. code-block:: bash
-
-   .vscode
-   ├── tools
-   │   ├── launch.template.json
-   │   ├── settings.template.json
-   │   └── setup_vscode.py
-   ├── extensions.json
-   ├── launch.json  # <- this is generated by setup_vscode.py
-   ├── settings.json  # <- this is generated by setup_vscode.py
-   └── tasks.json
-
-
-To setup the IDE, please follow these instructions:
-
-1. Open the ``Isaac Lab`` directory on Visual Studio Code IDE
-2. Run VSCode `Tasks <https://code.visualstudio.com/docs/editor/tasks>`__, by
-   pressing ``Ctrl+Shift+P``, selecting ``Tasks: Run Task`` and running the
-   ``setup_python_env`` in the drop down menu.
-
-   .. image:: ../_static/vscode_tasks.png
-      :width: 600px
-      :align: center
-      :alt: VSCode Tasks
-
-If everything executes correctly, it should create the following files:
-
-* ``.vscode/launch.json``: Contains the launch configurations for debugging python code.
-* ``.vscode/settings.json``: Contains the settings for the python interpreter and the python environment.
-
-For more information on VSCode support for Omniverse, please refer to the
-following links:
-
-* `Isaac Sim VSCode support <https://docs.omniverse.nvidia.com/app_isaacsim/app_isaacsim/manual_standalone_python.html#isaac-sim-python-vscode>`__
-* `Debugging with VSCode <https://docs.omniverse.nvidia.com/isaacsim/latest/advanced_tutorials/tutorial_advanced_python_debugging.html>`__
-
-
-Configuring the python interpreter
-----------------------------------
-
-In the provided configuration, we set the default python interpreter to use the
-python executable provided by Omniverse. This is specified in the
-``.vscode/settings.json`` file:
-
-.. code-block:: json
-
-   {
-      "python.defaultInterpreterPath": "${workspaceFolder}/_isaac_sim/python.sh",
-   }
-
-If you want to use a different python interpreter (for instance, from your conda environment),
-you need to change the python interpreter used by selecting and activating the python interpreter
-of your choice in the bottom left corner of VSCode, or opening the command palette (``Ctrl+Shift+P``)
-and selecting ``Python: Select Interpreter``.
-
-For more information on how to set python interpreter for VSCode, please
-refer to the `VSCode documentation <https://code.visualstudio.com/docs/python/environments#_working-with-python-interpreters>`_.
-
-Repository organization
------------------------
-
-The ``Isaac Lab`` repository is structured as follows:
-
-.. code-block:: bash
-
-   IsaacLab
-   ├── .vscode
-   ├── .flake8
-   ├── LICENSE
-   ├── isaaclab.sh
-   ├── pyproject.toml
-   ├── README.md
-   ├── docs
-   ├── source
-   │   ├── extensions
-   │   │   ├── omni.isaac.lab
-   │   │   ├── omni.isaac.lab_assets
-   │   │   └── omni.isaac.lab_tasks
-   │   ├── standalone
-   │   │   ├── demos
-   │   │   ├── environments
-   │   │   ├── tools
-   │   │   ├── tutorials
-   │   │   └── workflows
-   └── VERSION
-
-The ``source`` directory contains the source code for all ``Isaac Lab`` *extensions*
-and *standalone applications*. The two are the different development workflows
-supported in `Isaac Sim <https://docs.omniverse.nvidia.com/isaacsim/latest/introductory_tutorials/tutorial_intro_workflows.html>`__.
-These are described in the following sections.
-
 Extensions
 ~~~~~~~~~~
 

docs/source/overview/developer-guide/index.rst

+Developer's Guide
+=================
+
+For development, we suggest using `Microsoft Visual Studio Code
+(VSCode) <https://code.visualstudio.com/>`__. This is also suggested by
+NVIDIA Omniverse and there exists tutorials on how to `debug Omniverse
+extensions <https://www.youtube.com/watch?v=Vr1bLtF1f4U&ab_channel=NVIDIAOmniverse>`__
+using VSCode.
+
+.. toctree::
+  :maxdepth: 1
+
+  vs_code
+  repo_structure
+  extensions
+  template

docs/source/overview/developer-guide/repo_structure.rst

+Repository organization
+-----------------------
+
+The ``Isaac Lab`` repository is structured as follows:
+
+.. code-block:: bash
+
+   IsaacLab
+   ├── .vscode
+   ├── .flake8
+   ├── LICENSE
+   ├── isaaclab.sh
+   ├── pyproject.toml
+   ├── README.md
+   ├── docs
+   ├── source
+   │   ├── extensions
+   │   │   ├── omni.isaac.lab
+   │   │   ├── omni.isaac.lab_assets
+   │   │   └── omni.isaac.lab_tasks
+   │   ├── standalone
+   │   │   ├── demos
+   │   │   ├── environments
+   │   │   ├── tools
+   │   │   ├── tutorials
+   │   │   └── workflows
+   └── VERSION
+
+The ``source`` directory contains the source code for all ``Isaac Lab`` *extensions*
+and *standalone applications*. The two are the different development workflows
+supported in `Isaac Sim <https://docs.omniverse.nvidia.com/isaacsim/latest/introductory_tutorials/tutorial_intro_workflows.html>`__.
+These are described in the following sections.

docs/source/overview/developer-guide/vs_code.rst

+Setting up Visual Studio Code
+-----------------------------
+
+The following is only applicable for Isaac Sim installed via the Omniverse Launcher.
+The ``Isaac Lab`` repository includes the VSCode settings to easily allow setting
+up your development environment. These are included in the ``.vscode`` directory
+and include the following files:
+
+.. code-block:: bash
+
+   .vscode
+   ├── tools
+   │   ├── launch.template.json
+   │   ├── settings.template.json
+   │   └── setup_vscode.py
+   ├── extensions.json
+   ├── launch.json  # <- this is generated by setup_vscode.py
+   ├── settings.json  # <- this is generated by setup_vscode.py
+   └── tasks.json
+
+
+To setup the IDE, please follow these instructions:
+
+1. Open the ``Isaac Lab`` directory on Visual Studio Code IDE
+2. Run VSCode `Tasks <https://code.visualstudio.com/docs/editor/tasks>`__, by
+   pressing ``Ctrl+Shift+P``, selecting ``Tasks: Run Task`` and running the
+   ``setup_python_env`` in the drop down menu.
+
+   .. image:: ../../_static/vscode_tasks.png
+      :width: 600px
+      :align: center
+      :alt: VSCode Tasks
+
+If everything executes correctly, it should create the following files:
+
+* ``.vscode/launch.json``: Contains the launch configurations for debugging python code.
+* ``.vscode/settings.json``: Contains the settings for the python interpreter and the python environment.
+
+For more information on VSCode support for Omniverse, please refer to the
+following links:
+
+* `Isaac Sim VSCode support <https://docs.omniverse.nvidia.com/app_isaacsim/app_isaacsim/manual_standalone_python.html#isaac-sim-python-vscode>`__
+* `Debugging with VSCode <https://docs.omniverse.nvidia.com/isaacsim/latest/advanced_tutorials/tutorial_advanced_python_debugging.html>`__
+
+
+Configuring the python interpreter
+----------------------------------
+
+In the provided configuration, we set the default python interpreter to use the
+python executable provided by Omniverse. This is specified in the
+``.vscode/settings.json`` file:
+
+.. code-block:: json
+
+   {
+      "python.defaultInterpreterPath": "${workspaceFolder}/_isaac_sim/python.sh",
+   }
+
+If you want to use a different python interpreter (for instance, from your conda environment),
+you need to change the python interpreter used by selecting and activating the python interpreter
+of your choice in the bottom left corner of VSCode, or opening the command palette (``Ctrl+Shift+P``)
+and selecting ``Python: Select Interpreter``.
+
+For more information on how to set python interpreter for VSCode, please
+refer to the `VSCode documentation <https://code.visualstudio.com/docs/python/environments#_working-with-python-interpreters>`_.

docs/source/overview/environment-workflows/basic_environments.rst

+Environment Workflows
+=====================
+
+Workflows
+---------
+
+With Isaac Lab, we also provide a suite of benchmark environments included
+in the ``omni.isaac.lab_tasks`` extension. We use the OpenAI Gym registry
+to register these environments. For each environment, we provide a default
+configuration file that defines the scene, observations, rewards and action spaces.
+
+The list of environments available registered with OpenAI Gym can be found by running:
+
+.. code:: bash
+
+   ./isaaclab.sh -p source/standalone/environments/list_envs.py
+
+
+Basic agents
+~~~~~~~~~~~~
+
+These include basic agents that output zero or random agents. They are
+useful to ensure that the environments are configured correctly.
+
+-  Zero-action agent on the Cart-pole example
+
+   .. code:: bash
+
+      ./isaaclab.sh -p source/standalone/environments/zero_agent.py --task Isaac-Cartpole-v0 --num_envs 32
+
+-  Random-action agent on the Cart-pole example:
+
+   .. code:: bash
+
+      ./isaaclab.sh -p source/standalone/environments/random_agent.py --task Isaac-Cartpole-v0 --num_envs 32
+
+
+State machine
+~~~~~~~~~~~~~
+
+We include examples on hand-crafted state machines for the environments. These
+help in understanding the environment and how to use the provided interfaces.
+The state machines are written in `warp <https://github.com/NVIDIA/warp>`__ which
+allows efficient execution for large number of environments using CUDA kernels.
+
+.. code:: bash
+
+   ./isaaclab.sh -p source/standalone/environments/state_machine/lift_cube_sm.py --num_envs 32

docs/source/overview/environment-workflows/index.rst

+Introduction to Environments
+============================
+
+This section introduces fundamental concepts of environments and agents in Isaac Lab.
+Example scripts are available to show implementations of basic agents and state machines.
+In addition, we introduce the different task workflows available in Isaac Lab for
+implementing new environments.
+
+.. toctree::
+  :maxdepth: 1
+
+  basic_environments
+  task_workflows

docs/source/features/task_workflows.rst → docs/source/overview/environment-workflows/task_workflows.rst

@@ -68,7 +68,7 @@ for each component (such as the ``ObservationCfg`` and ``RewardCfg``).
     implementation, weight and additional parameters to be passed to the function. Users can define multiple
     reward terms and their weights to be used in the reward function.
 
-    .. literalinclude:: ../../../source/extensions/omni.isaac.lab_tasks/omni/isaac/lab_tasks/manager_based/classic/cartpole/cartpole_env_cfg.py
+    .. literalinclude:: ../../../../source/extensions/omni.isaac.lab_tasks/omni/isaac/lab_tasks/manager_based/classic/cartpole/cartpole_env_cfg.py
         :language: python
         :pyobject: RewardsCfg
 
@@ -105,14 +105,14 @@ for setting up the scene, processing the actions, computing the rewards, observa
 
     The following function is a part of the Cartpole environment class and is responsible for computing the rewards.
 
-    .. literalinclude:: ../../../source/extensions/omni.isaac.lab_tasks/omni/isaac/lab_tasks/direct/cartpole/cartpole_env.py
+    .. literalinclude:: ../../../../source/extensions/omni.isaac.lab_tasks/omni/isaac/lab_tasks/direct/cartpole/cartpole_env.py
         :language: python
         :pyobject: CartpoleEnv._get_rewards
         :dedent: 4
 
     It calls the :meth:`compute_rewards` function which is Torch JIT compiled for performance benefits.
 
-    .. literalinclude:: ../../../source/extensions/omni.isaac.lab_tasks/omni/isaac/lab_tasks/direct/cartpole/cartpole_env.py
+    .. literalinclude:: ../../../../source/extensions/omni.isaac.lab_tasks/omni/isaac/lab_tasks/direct/cartpole/cartpole_env.py
         :language: python
         :pyobject: compute_rewards
 

docs/source/overview/reinforcement-learning/index.rst

+Reinforcement Learning
+======================
+
+Isaac Lab supports multiple reinforcement learning frameworks.
+In this section, we show existing scripts for running reinforcement learning
+with supported RL libraries and provide a comparison of the supported
+learning frameworks.
+
+.. toctree::
+  :maxdepth: 1
+
+  rl_existing_scripts
+  rl_frameworks
+  performance_benchmarks

docs/source/overview/reinforcement-learning/performance_benchmarks.rst

+Performance Benchmarks
+======================
+
+Isaac Lab leverages end-to-end GPU training for reinforcement learning workflows,
+allowing for fast parallel training across thousands of environments.
+In this section, we provide runtime performance benchmark results for reinforcement learning
+training of various example environments on different GPU setups.
+Multi-GPU and multi-node training performance results are also outlined.
+
+
+Benchmark Results
+-----------------
+
+All benchmarking results were performed with the RL Games library with ``--headless`` flag on Ubuntu 22.04.
+``Isaac-Velocity-Rough-G1-v0`` environment benchmarks were performed with the RSL RL library.
+
+
+Memory Consumption
+^^^^^^^^^^^^^^^^^^
+
++-------------------------------------+-------------------+----------+-----------+
+| Environment Name                    | # of Environments | RAM (GB) | VRAM (GB) |
++=====================================+===================+==========+===========+
+| Isaac-Cartpole-Direct-v0            | 4096              | 3.7      | 3.3       |
++-------------------------------------+-------------------+----------+-----------+
+| Isaac-Cartpole-RGB-Camera-Direct-v0 | 1024              | 7.5      | 16.7      |
++-------------------------------------+-------------------+----------+-----------+
+| Isaac-Velocity-Rough-G1-v0          | 4096              | 6.5      | 6.1       |
++-------------------------------------+-------------------+----------+-----------+
+| Isaac-Repose-Cube-Shadow-Direct-v0  | 8192              | 6.7      | 6.4       |
++-------------------------------------+-------------------+----------+-----------+
+
+
+Single GPU - RTX 4090
+^^^^^^^^^^^^^^^^^^^^^
+
+CPU: AMD Ryzen 9 7950X 16-Core Processor
+
++-------------------------------------+-------------------+--------------+-------------------+--------------------+
+| Environment Name                    | # of Environments | Environment  | Environment Step  | Environment Step,  |
+|                                     |                   | Step FPS     | and               | Inference,         |
+|                                     |                   |              | Inference FPS     | and Train FPS      |
++=====================================+===================+==============+===================+====================+
+| Isaac-Cartpole-Direct-v0            | 4096              | 1100000      | 910000            | 510000             |
++-------------------------------------+-------------------+--------------+-------------------+--------------------+
+| Isaac-Cartpole-RGB-Camera-Direct-v0 | 1024              | 50000        | 45000             | 32000              |
++-------------------------------------+-------------------+--------------+-------------------+--------------------+
+| Isaac-Velocity-Rough-G1-v0          | 4096              | 94000        | 88000             | 82000              |
++-------------------------------------+-------------------+--------------+-------------------+--------------------+
+| Isaac-Repose-Cube-Shadow-Direct-v0  | 8192              | 200000       | 190000            | 170000             |
++-------------------------------------+-------------------+--------------+-------------------+--------------------+
+
+
+Single GPU - L40
+^^^^^^^^^^^^^^^^
+
+CPU: Intel(R) Xeon(R) Platinum 8362 CPU @ 2.80GHz
+
++-------------------------------------+-------------------+--------------+-------------------+--------------------+
+| Environment Name                    | # of Environments | Environment  | Environment Step  | Environment Step,  |
+|                                     |                   | Step FPS     | and               | Inference,         |
+|                                     |                   |              | Inference FPS     | and Train FPS      |
++=====================================+===================+==============+===================+====================+
+| Isaac-Cartpole-Direct-v0            | 4096              | 620000       | 490000            | 260000             |
++-------------------------------------+-------------------+--------------+-------------------+--------------------+
+| Isaac-Cartpole-RGB-Camera-Direct-v0 | 1024              | 30000        | 28000             | 21000              |
++-------------------------------------+-------------------+--------------+-------------------+--------------------+
+| Isaac-Velocity-Rough-G1-v0          | 4096              | 72000        | 64000             | 62000              |
++-------------------------------------+-------------------+--------------+-------------------+--------------------+
+| Isaac-Repose-Cube-Shadow-Direct-v0  | 8192              | 170000       | 140000            | 120000             |
++-------------------------------------+-------------------+--------------+-------------------+--------------------+
+
+
+Single-Node, 4 x L40 GPUs
+^^^^^^^^^^^^^^^^^^^^^^^^^
+
+CPU: Intel(R) Xeon(R) Platinum 8362 CPU @ 2.80GHz
+
++-------------------------------------+-------------------+--------------+-------------------+--------------------+
+| Environment Name                    | # of Environments | Environment  | Environment Step  | Environment Step,  |
+|                                     |                   | Step FPS     | and               | Inference,         |
+|                                     |                   |              | Inference FPS     | and Train FPS      |
++=====================================+===================+==============+===================+====================+
+| Isaac-Cartpole-Direct-v0            | 4096              | 2700000      | 2100000           | 950000             |
++-------------------------------------+-------------------+--------------+-------------------+--------------------+
+| Isaac-Cartpole-RGB-Camera-Direct-v0 | 1024              | 130000       | 120000            | 90000              |
++-------------------------------------+-------------------+--------------+-------------------+--------------------+
+| Isaac-Velocity-Rough-G1-v0          | 4096              | 290000       | 270000            | 250000             |
++-------------------------------------+-------------------+--------------+-------------------+--------------------+
+| Isaac-Repose-Cube-Shadow-Direct-v0  | 8192              | 440000       | 420000            | 390000             |
++-------------------------------------+-------------------+--------------+-------------------+--------------------+
+
+
+4 Nodes, 4 x L40 GPUs per node
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+CPU: Intel(R) Xeon(R) Platinum 8362 CPU @ 2.80GHz
+
++-------------------------------------+-------------------+--------------+-------------------+--------------------+
+| Environment Name                    | # of Environments | Environment  | Environment Step  | Environment Step,  |
+|                                     |                   | Step FPS     | and               | Inference,         |
+|                                     |                   |              | Inference FPS     | and Train FPS      |
++=====================================+===================+==============+===================+====================+
+| Isaac-Cartpole-Direct-v0            | 4096              | 10200000     | 8200000           | 3500000            |
++-------------------------------------+-------------------+--------------+-------------------+--------------------+
+| Isaac-Cartpole-RGB-Camera-Direct-v0 | 1024              | 530000       | 490000            | 260000             |
++-------------------------------------+-------------------+--------------+-------------------+--------------------+
+| Isaac-Velocity-Rough-G1-v0          | 4096              | 1200000      | 1100000           | 960000             |
++-------------------------------------+-------------------+--------------+-------------------+--------------------+
+| Isaac-Repose-Cube-Shadow-Direct-v0  | 8192              | 2400000      | 2300000           | 1800000            |
++-------------------------------------+-------------------+--------------+-------------------+--------------------+
+
+
+Benchmark Scripts
+-----------------
+
+For ease of reproducibility, we provide benchmarking scripts available at ``source/standalone/benchmarks``.
+This folder contains individual benchmark scripts that resemble the ``train.py`` script for RL-Games
+and RSL RL. In addition, we also provide a benchmarking script that runs only the environment implementation
+without any reinforcement learning library.
+
+Example scripts can be run similarly to training scripts:
+
+.. code-block:: bash
+
+   # benchmark with RSL RL
+   python source/standalone/benchmarks/benchmark_rsl_rl.py --task=Isaac-Cartpole-v0 --headless
+
+   # benchmark with RL Games
+   python source/standalone/benchmarks/benchmark_rlgames.py --task=Isaac-Cartpole-v0 --headless
+
+   # benchmark without RL libraries
+   python source/standalone/benchmarks/benchmark_non_rl.py --task=Isaac-Cartpole-v0 --headless
+
+Each script will generate a set of KPI files at the end of the run, which includes data on the
+startup times, runtime statistics, such as the time taken for each simulation or rendering step,
+as well as overall environment FPS for stepping the environment, performing inference during
+rollout, as well as training.

docs/source/setup/sample.rst → docs/source/overview/reinforcement-learning/rl_existing_scripts.rst

-Running Existing Scripts
-========================
-
-Showroom
---------
-
-The main core interface extension in Isaac Lab ``omni.isaac.lab`` provides
-the main modules for actuators, objects, robots and sensors. We provide
-a list of demo scripts and tutorials. These showcase how to use the provided
-interfaces within a code in a minimal way.
-
-A few quick showroom scripts to run and checkout:
-
--  Spawn different quadrupeds and make robots stand using position commands:
-
-   .. code:: bash
-
-      ./isaaclab.sh -p source/standalone/demos/quadrupeds.py
-
--  Spawn different arms and apply random joint position commands:
-
-   .. code:: bash
-
-      ./isaaclab.sh -p source/standalone/demos/arms.py
-
--  Spawn different hands and command them to open and close:
-
-   .. code:: bash
-
-      ./isaaclab.sh -p source/standalone/demos/hands.py
-
--  Spawn procedurally generated terrains with different configurations:
-
-   .. code:: bash
-
-      ./isaaclab.sh -p source/standalone/demos/procedural_terrain.py
-
--  Spawn different deformable (soft) bodies and let them fall from a height:
-
-   .. code:: bash
-
-      ./isaaclab.sh -p source/standalone/demos/deformables.py
-
--  Spawn multiple markers that are useful for visualizations:
-
-   .. code:: bash
-
-      ./isaaclab.sh -p source/standalone/demos/markers.py
-
-Workflows
----------
-
-With Isaac Lab, we also provide a suite of benchmark environments included
-in the ``omni.isaac.lab_tasks`` extension. We use the OpenAI Gym registry
-to register these environments. For each environment, we provide a default
-configuration file that defines the scene, observations, rewards and action spaces.
-
-The list of environments available registered with OpenAI Gym can be found by running:
-
-.. code:: bash
-
-   ./isaaclab.sh -p source/standalone/environments/list_envs.py
-
-
-Basic agents
-~~~~~~~~~~~~
-
-These include basic agents that output zero or random agents. They are
-useful to ensure that the environments are configured correctly.
-
--  Zero-action agent on the Cart-pole example
-
-   .. code:: bash
-
-      ./isaaclab.sh -p source/standalone/environments/zero_agent.py --task Isaac-Cartpole-v0 --num_envs 32
-
--  Random-action agent on the Cart-pole example:
-
-   .. code:: bash
-
-      ./isaaclab.sh -p source/standalone/environments/random_agent.py --task Isaac-Cartpole-v0 --num_envs 32
-
-
-State machine
-~~~~~~~~~~~~~
-
-We include examples on hand-crafted state machines for the environments. These
-help in understanding the environment and how to use the provided interfaces.
-The state machines are written in `warp <https://github.com/NVIDIA/warp>`__ which
-allows efficient execution for large number of environments using CUDA kernels.
-
-.. code:: bash
-
-   ./isaaclab.sh -p source/standalone/environments/state_machine/lift_cube_sm.py --num_envs 32
-
-
-Teleoperation
-~~~~~~~~~~~~~
-
-We provide interfaces for providing commands in SE(2) and SE(3) space
-for robot control. In case of SE(2) teleoperation, the returned command
-is the linear x-y velocity and yaw rate, while in SE(3), the returned
-command is a 6-D vector representing the change in pose.
-
-To play inverse kinematics (IK) control with a keyboard device:
-
-.. code:: bash
-
-   ./isaaclab.sh -p source/standalone/environments/teleoperation/teleop_se3_agent.py --task Isaac-Lift-Cube-Franka-IK-Rel-v0 --num_envs 1 --teleop_device keyboard
-
-The script prints the teleoperation events configured. For keyboard,
-these are as follows:
-
-.. code:: text
-
-   Keyboard Controller for SE(3): Se3Keyboard
-       Reset all commands: L
-       Toggle gripper (open/close): K
-       Move arm along x-axis: W/S
-       Move arm along y-axis: A/D
-       Move arm along z-axis: Q/E
-       Rotate arm along x-axis: Z/X
-       Rotate arm along y-axis: T/G
-       Rotate arm along z-axis: C/V
-
-Imitation Learning
-~~~~~~~~~~~~~~~~~~
-
-Using the teleoperation devices, it is also possible to collect data for
-learning from demonstrations (LfD). For this, we support the learning
-framework `Robomimic <https://robomimic.github.io/>`__ (Linux only) and allow saving
-data in
-`HDF5 <https://robomimic.github.io/docs/tutorials/dataset_contents.html#viewing-hdf5-dataset-structure>`__
-format.
-
-1. Collect demonstrations with teleoperation for the environment
-   ``Isaac-Lift-Cube-Franka-IK-Rel-v0``:
-
-   .. code:: bash
-
-      # step a: collect data with keyboard
-      ./isaaclab.sh -p source/standalone/workflows/robomimic/collect_demonstrations.py --task Isaac-Lift-Cube-Franka-IK-Rel-v0 --num_envs 1 --num_demos 10 --teleop_device keyboard
-      # step b: inspect the collected dataset
-      ./isaaclab.sh -p source/standalone/workflows/robomimic/tools/inspect_demonstrations.py logs/robomimic/Isaac-Lift-Cube-Franka-IK-Rel-v0/hdf_dataset.hdf5
-
-2. Split the dataset into train and validation set:
-
-   .. code:: bash
-
-      # install the dependencies
-      sudo apt install cmake build-essential
-      # install python module (for robomimic)
-      ./isaaclab.sh -i robomimic
-      # split data
-      ./isaaclab.sh -p source/standalone//workflows/robomimic/tools/split_train_val.py logs/robomimic/Isaac-Lift-Cube-Franka-IK-Rel-v0/hdf_dataset.hdf5 --ratio 0.2
-
-3. Train a BC agent for ``Isaac-Lift-Cube-Franka-IK-Rel-v0`` with
-   `Robomimic <https://robomimic.github.io/>`__:
-
-   .. code:: bash
-
-      ./isaaclab.sh -p source/standalone/workflows/robomimic/train.py --task Isaac-Lift-Cube-Franka-IK-Rel-v0 --algo bc --dataset logs/robomimic/Isaac-Lift-Cube-Franka-IK-Rel-v0/hdf_dataset.hdf5
-
-4. Play the learned model to visualize results:
-
-   .. code:: bash
-
-      ./isaaclab.sh -p source/standalone/workflows/robomimic/play.py --task Isaac-Lift-Cube-Franka-IK-Rel-v0 --checkpoint /PATH/TO/model.pth
-
-Reinforcement Learning
-~~~~~~~~~~~~~~~~~~~~~~
+Reinforcement Learning Wrappers
+===============================
 
 We provide wrappers to different reinforcement libraries. These wrappers convert the data
 from the environments into the respective libraries function argument and return types.
 
+Stable-Baselines3
+-----------------
+
 -  Training an agent with
    `Stable-Baselines3 <https://stable-baselines3.readthedocs.io/en/master/index.html>`__
    on ``Isaac-Cartpole-v0``:
@@ -189,6 +23,9 @@ from the environments into the respective libraries function argument and return
       # run script for recording video of a trained agent (requires installing `ffmpeg`)
       ./isaaclab.sh -p source/standalone/workflows/sb3/play.py --task Isaac-Cartpole-v0 --headless --video --video_length 200
 
+SKRL
+----
+
 -  Training an agent with
    `SKRL <https://skrl.readthedocs.io>`__ on ``Isaac-Reach-Franka-v0``:
 
@@ -222,6 +59,9 @@ from the environments into the respective libraries function argument and return
             # run script for recording video of a trained agent (requires installing `ffmpeg`)
             ./isaaclab.sh -p source/standalone/workflows/skrl/play.py --task Isaac-Reach-Franka-v0 --headless --ml_framework jax --video --video_length 200
 
+RL-Games
+--------
+
 -  Training an agent with
    `RL-Games <https://github.com/Denys88/rl_games>`__ on ``Isaac-Ant-v0``:
 
@@ -236,6 +76,10 @@ from the environments into the respective libraries function argument and return
       # run script for recording video of a trained agent (requires installing `ffmpeg`)
       ./isaaclab.sh -p source/standalone/workflows/rl_games/play.py --task Isaac-Ant-v0 --headless --video --video_length 200
 
+
+RSL-RL
+------
+
 -  Training an agent with
    `RSL-RL <https://github.com/leggedrobotics/rsl_rl>`__ on ``Isaac-Reach-Franka-v0``:
 

docs/source/overview/reinforcement-learning/rl_frameworks.rst

+Reinforcement Learning Library Comparison
+=========================================
+
+In this section, we provide an overview of the supported reinforcement learning libraries in Isaac Lab,
+along with performance benchmarks across the libraries.
+
+The supported libraries are:
+
+- `SKRL <https://skrl.readthedocs.io>`__
+- `RSL-RL <https://github.com/leggedrobotics/rsl_rl>`__
+- `RL-Games <https://github.com/Denys88/rl_games>`__
+- `Stable-Baselines3 <https://stable-baselines3.readthedocs.io/en/master/index.html>`__
+
+Feature Comparison
+------------------
+
+.. list-table::
+   :widths: 20 20 20 20 20
+   :header-rows: 1
+
+   * - Feature
+     - RL-Games
+     - RSL RL
+     - SKRL
+     - Stable Baselines3
+   * - Algorithms Included
+     - PPO, SAC, A2C
+     - PPO
+     - `Extensive List <https://skrl.readthedocs.io/en/latest/#agents>`__
+     - `Extensive List <https://github.com/DLR-RM/stable-baselines3?tab=readme-ov-file#implemented-algorithms>`__
+   * - Vectorized Training
+     - Yes
+     - Yes
+     - Yes
+     - No
+   * - Distributed Training
+     - Yes
+     - No
+     - Yes
+     - No
+   * - ML Frameworks Supported
+     - PyTorch
+     - PyTorch
+     - PyTorch, JAX
+     - PyTorch
+   * - Multi-Agent Support
+     - PPO
+     - PPO
+     - PPO + Multi-Agent algorithms
+     - External projects support
+   * - Documentation
+     - Low
+     - Low
+     - Moderate
+     - Extensive
+   * - Community Support
+     - Small Community
+     - Small Community
+     - Small Community
+     - Large Community
+   * - Available Examples in Isaac Lab
+     - Large
+     - Large
+     - Large
+     - Small
+
+
+Training Performance
+--------------------
+
+We performed training with each RL library on the same ``Isaac-Humanoid-v0`` environment
+with ``--headless`` on a single RTX 4090 GPU
+and logged the total training time for 65.5M steps for each RL library.
+
++--------------------+-----------------+
+| RL Library         | Time in seconds |
++====================+=================+
+| RL-Games           | 216             |
++--------------------+-----------------+
+| RSL RL             | 215             |
++--------------------+-----------------+
+| SKRL               | 321             |
++--------------------+-----------------+
+| Stable-Baselines3  | 6320            |
++--------------------+-----------------+

docs/source/overview/sample_scripts.rst

+Showroom Demos
+==============
+
+The main core interface extension in Isaac Lab ``omni.isaac.lab`` provides
+the main modules for actuators, objects, robots and sensors. We provide
+a list of demo scripts and tutorials. These showcase how to use the provided
+interfaces within a code in a minimal way.
+
+A few quick showroom scripts to run and checkout:
+
+-  Spawn different quadrupeds and make robots stand using position commands:
+
+   .. code:: bash
+
+      ./isaaclab.sh -p source/standalone/demos/quadrupeds.py
+
+-  Spawn different arms and apply random joint position commands:
+
+   .. code:: bash
+
+      ./isaaclab.sh -p source/standalone/demos/arms.py
+
+-  Spawn different hands and command them to open and close:
+
+   .. code:: bash
+
+      ./isaaclab.sh -p source/standalone/demos/hands.py
+
+-  Spawn procedurally generated terrains with different configurations:
+
+   .. code:: bash
+
+      ./isaaclab.sh -p source/standalone/demos/procedural_terrain.py
+
+-  Spawn different deformable (soft) bodies and let them fall from a height:
+
+   .. code:: bash
+
+      ./isaaclab.sh -p source/standalone/demos/deformables.py
+
+-  Spawn multiple markers that are useful for visualizations:
+
+   .. code:: bash
+
+      ./isaaclab.sh -p source/standalone/demos/markers.py

docs/source/overview/teleop_imitation.rst

+Teleoperation and Imitation Learning
+====================================
+
+
+Teleoperation
+~~~~~~~~~~~~~
+
+We provide interfaces for providing commands in SE(2) and SE(3) space
+for robot control. In case of SE(2) teleoperation, the returned command
+is the linear x-y velocity and yaw rate, while in SE(3), the returned
+command is a 6-D vector representing the change in pose.
+
+To play inverse kinematics (IK) control with a keyboard device:
+
+.. code:: bash
+
+   ./isaaclab.sh -p source/standalone/environments/teleoperation/teleop_se3_agent.py --task Isaac-Lift-Cube-Franka-IK-Rel-v0 --num_envs 1 --teleop_device keyboard
+
+The script prints the teleoperation events configured. For keyboard,
+these are as follows:
+
+.. code:: text
+
+   Keyboard Controller for SE(3): Se3Keyboard
+       Reset all commands: L
+       Toggle gripper (open/close): K
+       Move arm along x-axis: W/S
+       Move arm along y-axis: A/D
+       Move arm along z-axis: Q/E
+       Rotate arm along x-axis: Z/X
+       Rotate arm along y-axis: T/G
+       Rotate arm along z-axis: C/V
+
+Imitation Learning
+~~~~~~~~~~~~~~~~~~
+
+Using the teleoperation devices, it is also possible to collect data for
+learning from demonstrations (LfD). For this, we support the learning
+framework `Robomimic <https://robomimic.github.io/>`__ (Linux only) and allow saving
+data in
+`HDF5 <https://robomimic.github.io/docs/tutorials/dataset_contents.html#viewing-hdf5-dataset-structure>`__
+format.
+
+1. Collect demonstrations with teleoperation for the environment
+   ``Isaac-Lift-Cube-Franka-IK-Rel-v0``:
+
+   .. code:: bash
+
+      # step a: collect data with keyboard
+      ./isaaclab.sh -p source/standalone/workflows/robomimic/collect_demonstrations.py --task Isaac-Lift-Cube-Franka-IK-Rel-v0 --num_envs 1 --num_demos 10 --teleop_device keyboard
+      # step b: inspect the collected dataset
+      ./isaaclab.sh -p source/standalone/workflows/robomimic/tools/inspect_demonstrations.py logs/robomimic/Isaac-Lift-Cube-Franka-IK-Rel-v0/hdf_dataset.hdf5
+
+2. Split the dataset into train and validation set:
+
+   .. code:: bash
+
+      # install the dependencies
+      sudo apt install cmake build-essential
+      # install python module (for robomimic)
+      ./isaaclab.sh -i robomimic
+      # split data
+      ./isaaclab.sh -p source/standalone//workflows/robomimic/tools/split_train_val.py logs/robomimic/Isaac-Lift-Cube-Franka-IK-Rel-v0/hdf_dataset.hdf5 --ratio 0.2
+
+3. Train a BC agent for ``Isaac-Lift-Cube-Franka-IK-Rel-v0`` with
+   `Robomimic <https://robomimic.github.io/>`__:
+
+   .. code:: bash
+
+      ./isaaclab.sh -p source/standalone/workflows/robomimic/train.py --task Isaac-Lift-Cube-Franka-IK-Rel-v0 --algo bc --dataset logs/robomimic/Isaac-Lift-Cube-Franka-IK-Rel-v0/hdf_dataset.hdf5
+
+4. Play the learned model to visualize results:
+
+   .. code:: bash
+
+      ./isaaclab.sh -p source/standalone/workflows/robomimic/play.py --task Isaac-Lift-Cube-Franka-IK-Rel-v0 --checkpoint /PATH/TO/model.pth

source/extensions/omni.isaac.lab/omni/isaac/lab/sim/simulation_cfg.py

@@ -64,7 +64,7 @@ class PhysxCfg:
     """
 
     min_velocity_iteration_count: int = 0
-    """Minimum number of solver position iterations (rigid bodies, cloth, particles etc.). Default is 0.
+    """Minimum number of solver velocity iterations (rigid bodies, cloth, particles etc.). Default is 0.
 
     .. note::
 
@@ -74,7 +74,7 @@ class PhysxCfg:
     """
 
     max_velocity_iteration_count: int = 255
-    """Maximum number of solver position iterations (rigid bodies, cloth, particles etc.). Default is 255.
+    """Maximum number of solver velocity iterations (rigid bodies, cloth, particles etc.). Default is 255.
 
     .. note::
 

source/extensions/omni.isaac.lab_tasks/omni/isaac/lab_tasks/utils/wrappers/rl_games.py

@@ -3,7 +3,7 @@
 #
 # SPDX-License-Identifier: BSD-3-Clause
 
-"""Wrapper to configure an :class:`ManagerBasedRLEnv` instance to RL-Games vectorized environment.
+"""Wrapper to configure a :class:`ManagerBasedRLEnv` or :class:`DirectRlEnv` instance to RL-Games vectorized environment.
 
 The following example shows how to wrap an environment for RL-Games and register the environment construction
 for RL-Games :class:`Runner` class:
@@ -60,7 +60,7 @@ class RlGamesVecEnvWrapper(IVecEnv):
     observations. This dictionary contains "obs" and "states" which typically correspond
     to the actor and critic observations respectively.
 
-    To use asymmetric actor-critic, the environment observations from :class:`ManagerBasedRLEnv`
+    To use asymmetric actor-critic, the environment observations from :class:`ManagerBasedRLEnv` or :class:`DirectRLEnv`
     must have the key or group name "critic". The observation group is used to set the
     :attr:`num_states` (int) and :attr:`state_space` (:obj:`gym.spaces.Box`). These are
     used by the learning agent in RL-Games to allocate buffers in the trajectory memory.
@@ -79,7 +79,7 @@ class RlGamesVecEnvWrapper(IVecEnv):
         https://github.com/NVIDIA-Omniverse/IsaacGymEnvs
     """
 
-    def __init__(self, env: ManagerBasedRLEnv, rl_device: str, clip_obs: float, clip_actions: float):
+    def __init__(self, env: ManagerBasedRLEnv | DirectRLEnv, rl_device: str, clip_obs: float, clip_actions: float):
         """Initializes the wrapper instance.
 
         Args:
@@ -89,12 +89,15 @@ class RlGamesVecEnvWrapper(IVecEnv):
             clip_actions: The clipping value for actions.
 
         Raises:
-            ValueError: The environment is not inherited from :class:`ManagerBasedRLEnv`.
+            ValueError: The environment is not inherited from :class:`ManagerBasedRLEnv` or :class:`DirectRLEnv`.
             ValueError: If specified, the privileged observations (critic) are not of type :obj:`gym.spaces.Box`.
         """
         # check that input is valid
         if not isinstance(env.unwrapped, ManagerBasedRLEnv) and not isinstance(env.unwrapped, DirectRLEnv):
-            raise ValueError(f"The environment must be inherited from ManagerBasedRLEnv. Environment type: {type(env)}")
+            raise ValueError(
+                "The environment must be inherited from ManagerBasedRLEnv or DirectRLEnv. Environment type:"
+                f" {type(env)}"
+            )
         # initialize the wrapper
         self.env = env
         # store provided arguments
@@ -168,7 +171,7 @@ class RlGamesVecEnvWrapper(IVecEnv):
         return cls.__name__
 
     @property
-    def unwrapped(self) -> ManagerBasedRLEnv:
+    def unwrapped(self) -> ManagerBasedRLEnv | DirectRLEnv:
         """Returns the base environment of the wrapper.
 
         This will be the bare :class:`gymnasium.Env` environment, underneath all layers of wrappers.

source/extensions/omni.isaac.lab_tasks/omni/isaac/lab_tasks/utils/wrappers/rsl_rl/vecenv_wrapper.py

@@ -3,7 +3,7 @@
 #
 # SPDX-License-Identifier: BSD-3-Clause
 
-"""Wrapper to configure an :class:`ManagerBasedRLEnv` instance to RSL-RL vectorized environment.
+"""Wrapper to configure a :class:`ManagerBasedRLEnv` or :class:`DirectRlEnv` instance to RSL-RL vectorized environment.
 
 The following example shows how to wrap an environment for RSL-RL:
 
@@ -43,7 +43,7 @@ class RslRlVecEnvWrapper(VecEnv):
         https://github.com/leggedrobotics/rsl_rl/blob/master/rsl_rl/env/vec_env.py
     """
 
-    def __init__(self, env: ManagerBasedRLEnv):
+    def __init__(self, env: ManagerBasedRLEnv | DirectRLEnv):
         """Initializes the wrapper.
 
         Note:
@@ -53,7 +53,7 @@ class RslRlVecEnvWrapper(VecEnv):
             env: The environment to wrap around.
 
         Raises:
-            ValueError: When the environment is not an instance of :class:`ManagerBasedRLEnv`.
+            ValueError: When the environment is not an instance of :class:`ManagerBasedRLEnv` or :class:`DirectRLEnv`.
         """
         # check that input is valid
         if not isinstance(env.unwrapped, ManagerBasedRLEnv) and not isinstance(env.unwrapped, DirectRLEnv):
@@ -126,7 +126,7 @@ class RslRlVecEnvWrapper(VecEnv):
         return cls.__name__
 
     @property
-    def unwrapped(self) -> ManagerBasedRLEnv:
+    def unwrapped(self) -> ManagerBasedRLEnv | DirectRLEnv:
         """Returns the base environment of the wrapper.
 
         This will be the bare :class:`gymnasium.Env` environment, underneath all layers of wrappers.

source/extensions/omni.isaac.lab_tasks/omni/isaac/lab_tasks/utils/wrappers/sb3.py

@@ -3,7 +3,7 @@
 #
 # SPDX-License-Identifier: BSD-3-Clause
 
-"""Wrapper to configure an :class:`ManagerBasedRLEnv` instance to Stable-Baselines3 vectorized environment.
+"""Wrapper to configure a :class:`ManagerBasedRLEnv` or :class:`DirectRLEnv` instance to Stable-Baselines3 vectorized environment.
 
 The following example shows how to wrap an environment for Stable-Baselines3:
 
@@ -85,7 +85,7 @@ class Sb3VecEnvWrapper(VecEnv):
     still considered a single environment instance, Stable Baselines tries to wrap
     around it using the :class:`DummyVecEnv`. This is only done if the environment
     is not inheriting from their :class:`VecEnv`. Thus, this class thinly wraps
-    over the environment from :class:`ManagerBasedRLEnv`.
+    over the environment from :class:`ManagerBasedRLEnv` or :class:`DirectRLEnv`.
 
     Note:
         While Stable-Baselines3 supports Gym 0.26+ API, their vectorized environment
@@ -123,14 +123,14 @@ class Sb3VecEnvWrapper(VecEnv):
 
     """
 
-    def __init__(self, env: ManagerBasedRLEnv):
+    def __init__(self, env: ManagerBasedRLEnv | DirectRLEnv):
         """Initialize the wrapper.
 
         Args:
             env: The environment to wrap around.
 
         Raises:
-            ValueError: When the environment is not an instance of :class:`ManagerBasedRLEnv`.
+            ValueError: When the environment is not an instance of :class:`ManagerBasedRLEnv` or :class:`DirectRLEnv`.
         """
         # check that input is valid
         if not isinstance(env.unwrapped, ManagerBasedRLEnv) and not isinstance(env.unwrapped, DirectRLEnv):
@@ -177,7 +177,7 @@ class Sb3VecEnvWrapper(VecEnv):
         return cls.__name__
 
     @property
-    def unwrapped(self) -> ManagerBasedRLEnv:
+    def unwrapped(self) -> ManagerBasedRLEnv | DirectRLEnv:
         """Returns the base environment of the wrapper.
 
         This will be the bare :class:`gymnasium.Env` environment, underneath all layers of wrappers.

source/extensions/omni.isaac.lab_tasks/omni/isaac/lab_tasks/utils/wrappers/skrl.py

@@ -3,7 +3,7 @@
 #
 # SPDX-License-Identifier: BSD-3-Clause
 
-"""Wrapper to configure an :class:`ManagerBasedRLEnv` instance to skrl environment.
+"""Wrapper to configure a :class:`ManagerBasedRLEnv` or :class:`DirectRLEnv` instance to skrl environment.
 
 The following example shows how to wrap an environment for skrl:
 
@@ -100,10 +100,12 @@ Vectorized environment wrapper.
 """
 
 
-def SkrlVecEnvWrapper(env: ManagerBasedRLEnv, ml_framework: Literal["torch", "jax", "jax-numpy"] = "torch"):
+def SkrlVecEnvWrapper(
+    env: ManagerBasedRLEnv | DirectRLEnv, ml_framework: Literal["torch", "jax", "jax-numpy"] = "torch"
+):
     """Wraps around Isaac Lab environment for skrl.
 
-    This function wraps around the Isaac Lab environment. Since the :class:`ManagerBasedRLEnv` environment
+    This function wraps around the Isaac Lab environment. Since the :class:`ManagerBasedRLEnv` or :class:`DirectRLEnv` environment
     wrapping functionality is defined within the skrl library itself, this implementation
     is maintained for compatibility with the structure of the extension that contains it.
     Internally it calls the :func:`wrap_env` from the skrl library API.
@@ -113,7 +115,7 @@ def SkrlVecEnvWrapper(env: ManagerBasedRLEnv, ml_framework: Literal["torch", "ja
         ml_framework: The ML framework to use for the wrapper. Defaults to "torch".
 
     Raises:
-        ValueError: When the environment is not an instance of :class:`ManagerBasedRLEnv`.
+        ValueError: When the environment is not an instance of :class:`ManagerBasedRLEnv` or :class:`DirectRLEnv`.
         ValueError: If the specified ML framework is not valid.
 
     Reference: