Data

Tools

Indexes

Applications

Experiments

Open Source Science

Recent Releases of gymnasium-robotics

gymnasium-robotics - 1.4.1

  • Fixes the environment overwrite warnings raised for the Mujoco v2 and v3 environments added in 1.4.0

  • Fixes the website build process

Full Changelog: https://github.com/Farama-Foundation/Gymnasium-Robotics/compare/v1.4.0...v1.4.1

- Python
Published by pseudo-rnd-thoughts 7 months ago

gymnasium-robotics - 1.4

Release Notes

Highlights

This release focuses on updating dependencies, refining Python version support, and migrating legacy MuJoCo environments to a new package.

What's Changed

  • Dependency Update: gymnasium is now required to be version 1.2 or higher.
  • Python Support:
    • Added support for Python 3.13.
    • Dropped support for Python 3.8 and 3.9.
  • MuJoCo Environments v2 & v3 Migration: The Gymnasium/MuJoCo environments (v2 and v3), which were based on the legacy mujoco_py library, have been moved to the Gymnasium-Robotics package. This move was completed in pull request https://github.com/Farama-Foundation/Gymnasium-Robotics/pull/271 @Kallinteris-Andreas and @pseudo-rnd-thoughts

For Users of Legacy mujoco-py Environments:

To ensure compatibility, the legacy environments now rely on the mujoco-py-cython3 package. You can install the necessary extras with:

bash pip install gymnasium-robotics[mujoco_py] If you need to use older mujoco-py versions for your work, you can now use the mujoco-py-original extra: bash pip install gymnasium-robotics[mujoco_py_original]

- Python
Published by Kallinteris-Andreas 8 months ago

gymnasium-robotics - 1.3.2

Release Notes:

Added Fetch*-v4 and HandReach-v3 environments, fixing the state initialization bug found in the previous version

And fixed a few bugs

What's Changed

Bug Fixes:

  • Fix bugs in kettle_asset.xml to ensure compatibility with the latest MuJoCo versions https://github.com/Farama-Foundation/Gymnasium-Robotics/pull/255 @Josh00-Lu (validation by @Kallinteris-Andreas)
  • Fixes Fetch and Robotic environments initial state issues in 1.3.0 https://github.com/Farama-Foundation/Gymnasium-Robotics/pull/256 @wjxgeorge
  • Fix rgba value in oven_asset.xml to be in [0,1] https://github.com/Farama-Foundation/Gymnasium-Robotics/pull/262 @RussTedrake

Documentation Updates:

  • set copyright year in documentation page automatically https://github.com/Farama-Foundation/Gymnasium-Robotics/pull/249 @Kallinteris-Andreas
  • typo fixes MaMuJoCo changelog https://github.com/Farama-Foundation/Gymnasium-Robotics/pull/250 @Kallinteris-Andreas
  • Fixed reward description of PointMaze dense environments https://github.com/Farama-Foundation/Gymnasium-Robotics/pull/253 @HridayM25
  • Fix Documentation toctrees menus https://github.com/Farama-Foundation/Gymnasium-Robotics/pull/254 @Kallinteris-Andreas @younik
  • Make logo image in README.md link to documentation https://github.com/Farama-Foundation/Gymnasium-Robotics/pull/261@Tanmay692004

Full Changelog: https://github.com/Farama-Foundation/Gymnasium-Robotics/compare/v1.3.1...v1.3.2

- Python
Published by Kallinteris-Andreas 9 months ago

gymnasium-robotics - 1.3.1

Release Notes:

A few bug fixes and fixes the internal testing

Bug Fixes:

  • Fix rendering bug by setting frame height and width https://github.com/Farama-Foundation/Gymnasium-Robotics/pull/236 @violasox
  • Re-enable disabled test_envs.py (mujoco only) https://github.com/Farama-Foundation/Gymnasium-Robotics/pull/243 @Kallinteris-Andreas
  • Re-enable environment specific tests https://github.com/Farama-Foundation/Gymnasium-Robotics/pull/247 @Kallinteris-Andreas
  • Fix MazeAnt-v5 environments having incorrect entry points https://github.com/Farama-Foundation/Gymnasium-Robotics/pull/240 @Kallinteris-Andreas

Minor Changes:

  • Update pre-commit https://github.com/Farama-Foundation/Gymnasium-Robotics/pull/246 @Kallinteris-Andreas
  • Use pipx for CI pre-commit https://github.com/Farama-Foundation/Gymnasium-Robotics/pull/241 @Kallinteris-Andreas

Documentation Updates:

  • update MaMuJoCo doc "quadruped|arm" by @Kallinteris-Andreas in https://github.com/Farama-Foundation/Gymnasium-Robotics/pull/244

New Contributors

  • @violasox made their first contribution in https://github.com/Farama-Foundation/Gymnasium-Robotics/pull/236

Full Changelog: https://github.com/Farama-Foundation/Gymnasium-Robotics/compare/v1.3.0...v1.3.1

- Python
Published by Kallinteris-Andreas over 1 year ago

gymnasium-robotics - v1.3

Release Notes:

1.3 Is a Major Release, adding new versions of the environments and supporting gymnasium==1.0.0

New Environments:

  • Add MaMuJoCo-v1 Environments @Kallinteris-Andreas (https://github.com/Farama-Foundation/Gymnasium-Robotics/pull/196)
  • Added Fetch-v3 and HandReach-v2 environments that fix reproducibility issues @amacati (https://github.com/Farama-Foundation/Gymnasium-Robotics/pull/208/)
  • Add AntMaze_*-v5 environments, featuring support for loading Third party robot models to Maze @Kallinteris-Andreas (https://github.com/Farama-Foundation/Gymnasium-Robotics/pull/217) <!-- ## What's Changed -->

MaMuJoCo-V1

  • Now based on Gymnasium/MuJoCo-v5 instead of Gymnasium/MuJoCo-v4 (https://github.com/Farama-Foundation/Gymnasium/pull/572).
  • When factorizatoion=None, the env.gent_action_partitions.dummy_node now contains action_id (it used to be None).
  • Added map_local_observations_to_global_state & optimized runtime performance of map_global_state_to_local_observations.
  • Added gym_env argument, which can be used to load third-party Gymansium.MujocoEnv environments.

Ant

  • Now observes local_categories of cfrc_ext by default (same as Gymnasium/MuJoCo-v5/Ant).
  • Renamed global node torso → root.

Humanoid(-Standup)

  • No longer observes qfrc_actuator of root & cinert, cvel, qfrc_actuator, cfrc_ext of worldbody (same as Gymnasium/MuJoCo-v5/Humanoid(-Standup)).

Walker2d

  • Fixed bug: global nodes are now [root_x, root_z, root_y] (used to be [root_x, root_x, root_z]).

ManySegmentAnt

  • frame_skip default set to 5 (same as Gymnasium/Ant).
  • option.timestep set to 0.01 (same as Gymnasium/Ant).
  • Now uses the same reward function as Gymnasium/Ant.
  • Now observes cfrc_ext by default, (same as Gymnasium/MuJoCo-v5).

ManySegmentSwimmer

  • Now uses the same option.timestep as Gymansum/Swimmer (0.01).
  • Updated model to work with mujoco>=3.0.0.

Breaking changes:

  • Now supports and requires gymnasium>=1.0 @Kallinteris-Andreas (https://github.com/Farama-Foundation/Gymnasium-Robotics/pull/211)

Bug Fixes:

Dependency Updates:

  • Added support for python==3.12 @Kallinteris-Andreas (https://github.com/Farama-Foundation/Gymnasium-Robotics/pull/231)

Minor Changes:

  • [adroit environments] Minor Cleanup: assignment of goal_achieved @Kallinteris-Andreas (https://github.com/Farama-Foundation/Gymnasium-Robotics/pull/205)
  • Add asserts for mj_id2name calls in mujoco_utils @DavidPL1 https://github.com/Farama-Foundation/Gymnasium-Robotics/pull/218

Documentation Updates:

  • minor typo fix https://github.com/Farama-Foundation/Gymnasium-Robotics/pull/221 @timofriedl
  • many Documentation updates and fixes by @Kallinteris-Andreas

New Contributors

  • @amacati made their first contribution in https://github.com/Farama-Foundation/Gymnasium-Robotics/pull/208
  • @DavidPL1 made their first contribution in https://github.com/Farama-Foundation/Gymnasium-Robotics/pull/218
  • @timofriedl made their first contribution in https://github.com/Farama-Foundation/Gymnasium-Robotics/pull/221

Full Changelog: https://github.com/Farama-Foundation/Gymnasium-Robotics/compare/v1.2.4...v1.3.0

- Python
Published by Kallinteris-Andreas over 1 year ago

gymnasium-robotics - v1.2.4

Release Notes:

A minor release bringing a bug fix and registering and pre-existing environment. This is the first release since @Kallinteris-Andreas became the project manager.

What's Changed

New Features:

  • Register PointMaze_Open_Diverse_GR by @dtch1997 in https://github.com/Farama-Foundation/Gymnasium-Robotics/pull/195

Bug Fixes:

  • randomize maze environments temporary xml file name by @Kallinteris-Andreas in https://github.com/Farama-Foundation/Gymnasium-Robotics/pull/185

Dependency Updates:

  • limit mujoco<3.0 by @Kallinteris-Andreas in https://github.com/Farama-Foundation/Gymnasium-Robotics/pull/187

Minor Changes:

  • Update to pyright==1.1.339 by @Kallinteris-Andreas in https://github.com/Farama-Foundation/Gymnasium-Robotics/pull/191
  • Update pyproject.toml python version to 3.8 by @Kallinteris-Andreas in https://github.com/Farama-Foundation/Gymnasium-Robotics/pull/191

Documentation Updates:

  • update observation space docstring: obs[6:8] is actually block - gripper for fetch pick_and_place push slide environments by @SethPate in https://github.com/Farama-Foundation/Gymnasium-Robotics/pull/197
  • remove py3.7 from installation.md documentation by @Kallinteris-Andreas in https://github.com/Farama-Foundation/Gymnasium-Robotics/pull/199

Full Changelog: https://github.com/Farama-Foundation/Gymnasium-Robotics/compare/v1.2.3...v1.2.4

- Python
Published by Kallinteris-Andreas about 2 years ago

gymnasium-robotics - v1.2.3

Gymnasium-Robotics v1.2.3 Release Notes:

Breaking changes:

  • Drop support for Python 3.7 which has reached its end of life. (#159)
  • New v4 version for the AntMaze environments that fix the following issue #155. (#156)

Bug Fixes:

  • Allow to compute rewards from batched observations in maze environments (PointMaze/AntMaze) (#153, #158)
  • Bump AntMaze environments version to v4 which fix issue #155. The following new files have been added to the source code: ant_maze_v4.py and maze_v4.py. (#156). The fixes involve:
    • When the environment is initialized with continuing_task=True, the reward is now calculated before resetting the goal location. Previously the reward was always zero whether the ant reached the goal or not during the full episode.
    • Fix the ant agent being reset into a terminal state. The maze_size_scaling factor was missing in the distance check in MazeEnv.generate_reset_pos().
    • Add success item to info return. info["success"].
  • Fix goal_cell and reset_cell assertions when reset maze environments (#164, #179)
  • Fix issue #166 in FrankaKitchen environment. info["tasks_to_complete"] was not giving the correct values. (#169)

New Features

  • Add reset_target boolean argument for initializing maze environments. If reset_target=True and continuing_task=True, the goal will be automatically placed in a new location when the agent reaches it in the same episode. If reset_target=False and continuing_task=True, the goal location won't be updated when reached by the agent and reward will be accumulated as long as the agent stays around the goal threshold. (#167, #170)
  • For maze environments, if the goal and reset cell locations are not given in the maze map structuree, they will be chosen automatically among empty cells. (#170)

Dependency Updates

  • Remove restrictions on numpy version, numpy>=1.21.0 (#154)
  • Remove restrictions on mujoco version, mujoco>=2.3.3 (#171)
  • Restrict cython version to cython<3 due to the following issue https://github.com/Farama-Foundation/Gymnasium/pull/616, (#162)

Documentation Updates

  • Replace main logo svg format with png (#160)
  • Update sphinx to latest version (#157)
  • Add release notes changelog (#174)
  • Remove versioning for included environments in documentation and update gifs for maze environments (#172, #177)
  • Fix table format for Shadow Dexterous Hand - Reach environment (#178)

Full Changelog: https://github.com/Farama-Foundation/Gymnasium-Robotics/compare/v1.2.2...v1.2.3

- Python
Published by rodrigodelazcano over 2 years ago

gymnasium-robotics - v1.2.2

Release Notes

This minor release updates MaMuJoCo to follow the latest PettingZoo version 1.23.0 and some minor bug fixes in the Github PyPI publish workflow.

New Features

  • Update MaMujoco to PettingZoo 1.23.0 by @Kallinteris-Andreas in https://github.com/Farama-Foundation/Gymnasium-Robotics/pull/150 ### Bug Fix
  • Include Franka mesh files in the setuptools.package-data parameter of pyproject.toml by @rodrigodelazcano 9db9196a0d1a927fa61443d7c6b906a10804b1ec

Documentation Updates

  • MaMuJoCo documentation, remove install from source instructions by @Kallinteris-Andreas in https://github.com/Farama-Foundation/Gymnasium-Robotics/pull/151

Full Changelog: https://github.com/Farama-Foundation/Gymnasium-Robotics/compare/v1.2.1...v1.2.2

- Python
Published by rodrigodelazcano almost 3 years ago

gymnasium-robotics - v1.2.1

Gymnasium-Robotics 1.2.1 Release Notes:

This minor release adds new Multi-agent environments from the MaMuJoCo project. These environments have been updated to follow the PettingZoo API and use the latest mujoco bindings. In addition, the updates made for the first release of FrankaKitchen-v1 environment have been reverted in order for the environment to resemble more its original version in relay-policy-learning and D4RL. This will solve existing confusion with the action space (https://github.com/Farama-Foundation/Gymnasium-Robotics/issues/135) and facilitate the re-creation of datasets in Minari.

We are also pining the mujoco version to v2.3.3 until we address the following issue (https://github.com/deepmind/mujoco/issues/833).

Breaking Changes

  • Revert FrankaKitchen-v1 environment to original. @rodrigodelazcano in https://github.com/Farama-Foundation/Gymnasium-Robotics/pull/145. These changes involve:
    • robot model: use the Franka robot model of the original environment instead of the model provided in mujoco_menagerie
    • action space: remove the Inverse Kinematic control option and maintain a single action space, the original joint velocity control.
    • goal tasks: some task have been removed which were not present in the original environment (toprightburner and bottomrightburner). Also the tasks name now match the original naming.

New Features

  • Add MaMuJoCo (Multi-agent mujoco) environments by @Kallinteris-Andreas in https://github.com/Farama-Foundation/Gymnasium-Robotics/pull/53. Documentation has been also included at https://robotics.farama.org/envs/MaMuJoCo/. NOTE: we are currently in the process of validating this environments https://github.com/Farama-Foundation/Gymnasium-Robotics/issues/141
  • Initialize PointMaze and AntMaze environments with random goal and reset position by default . @rodrigodelazcano in https://github.com/Farama-Foundation/Gymnasium-Robotics/pull/110, https://github.com/Farama-Foundation/Gymnasium-Robotics/pull/114
  • Add success key to infos return dictionary in all Maze environments. @rodrigodelazcano in https://github.com/Farama-Foundation/Gymnasium-Robotics/pull/110
  • Recover the set_env_state(state_dict={}) method of the Adroit hand environments from https://github.com/vikashplus/mjenvs . The initial state of the simulation can also be set by passing the dictionary argument `initialstatedictwhen callingenv.reset(options={'initialstate_dict': Dict)`. @rodrigodelazcano in https://github.com/Farama-Foundation/Gymnasium-Robotics/pull/119, @rodrigodelazcano in https://github.com/Farama-Foundation/Gymnasium-Robotics/pull/115
  • Resparsify adroit hand envs by @jjshoots in https://github.com/Farama-Foundation/Gymnasium-Robotics/pull/111

Bug Fixes

  • Add missing underscore to fix rendering by @frankroeder in https://github.com/Farama-Foundation/Gymnasium-Robotics/pull/102
  • Correct point_obs slicing for achieved_goal in PointMaze environments by @dohmjan in https://github.com/Farama-Foundation/Gymnasium-Robotics/pull/105
  • Update the position of the goal every reset in AntMaze environment by @nicehiro in https://github.com/Farama-Foundation/Gymnasium-Robotics/pull/106
  • Correct FetchReach environment versioning from v3 to v2 by @aalmuzairee in https://github.com/Farama-Foundation/Gymnasium-Robotics/pull/121
  • Fix issue #128 . Use jnt_doafdr instead of nt_qposadr for the mujoco_utils.get_joint_qvel() utility function. by @rodrigodelazcano in https://github.com/Farama-Foundation/Gymnasium-Robotics/pull/129
  • Correct x, y scaling for Maze environments @rodrigodelazcano in https://github.com/Farama-Foundation/Gymnasium-Robotics/pull/110
  • Fix door state space key by @rodrigodelazcano in https://github.com/Farama-Foundation/Gymnasium-Robotics/pull/130
  • Make getter functions for qpos / qvel return copies by @hueds in https://github.com/Farama-Foundation/Gymnasium-Robotics/pull/136

Minor Changes

  • Enable pyright.reportOptionalMemberAccess by @Kallinteris-Andreas in https://github.com/Farama-Foundation/Gymnasium-Robotics/pull/93
  • Add Farama Notifications by @jjshoots in https://github.com/Farama-Foundation/Gymnasium-Robotics/pull/120

Documentation

  • Fix observation space table in FetchSlide docs. @rodrigodelazcano in https://github.com/Farama-Foundation/Gymnasium-Robotics/pull/109
  • Update docs/README.md to link to a new CONTRIBUTING.md for docs by @mgoulao in https://github.com/Farama-Foundation/Gymnasium-Robotics/pull/117
  • Add docs versioning and release notes by @mgoulao in https://github.com/Farama-Foundation/Gymnasium-Robotics/pull/124
  • Fix missing edit button by @mgoulao in https://github.com/Farama-Foundation/Gymnasium-Robotics/pull/138
  • Add missing docs requirement by @mgoulao in https://github.com/Farama-Foundation/Gymnasium-Robotics/pull/125
  • Add sparse reward variant for AdroitHand environments by @jjshoots in https://github.com/Farama-Foundation/Gymnasium-Robotics/pull/123

Full Changelog: https://github.com/Farama-Foundation/Gymnasium-Robotics/compare/v1.2.0...v1.2.1

- Python
Published by rodrigodelazcano almost 3 years ago

gymnasium-robotics - Version 1.2.0

Finally here! :partying_face: :robot:

Refactored versions of the D4RL MuJoCo environments are now available in Gymnasium-Robotics (PointMaze, AntMaze, AdroitHand, and FrankaKitchen). The configuration of these environments is not identically to the originals, please read the new details in the documentation webpage at https://robotics.farama.org/

Moving forward, we are recreating the offline datasets with Minari and evaluating the environments. If you have any questions or would like to contribute please don't hesitate to reach out to us through the following discord channel. https://discord.com/channels/961771112864313344/1017088934238498837

What's Changed

  • Add three different refactored environment types from MuJoCo D4RL and update to the Gymnasium API standards version 0.27.0. Point Maze, Ant Maze, Adroit Hand, and FrankaKitchen @rodrigodelazcano
  • Add sparse reward option to the Adroit Hand environments @jjshoots in #69
  • Standardize file tree structure to facilitate automatic documentation generation. The environment file structure should look as follows: gymnasium_robotics.envs.env_type.env_name:EnvClass. @rodrigodelazcano in #83
  • Update to new Gymnasium MujocoRenderer class https://github.com/Farama-Foundation/Gymnasium/pull/112 for rendering @rodrigodelazcano
  • Add pydocstyle to pre-commit @rodrigodelazcano

Other contributions

  • Miscellaneous documentation webpage fixes @mgoulao, @SiddarGu, @jjshoots
  • Pin numpy to numpy<1.24.0 due to #221 @rodrigodelazcano
  • Move dependency installs and setuptools to pyproject.toml. Remove requirements.txt and test_requirements.txt @jjshoots
  • Add google analytics to webpage @mgoulao
  • Switch flake8 from gitlab to github @RedTachyon in https://github.com/Farama-Foundation/Gymnasium-Robotics/pull/52
  • Code/Documentation typo fixes @araffin , @Kallinteris-Andreas

New Contributors

  • @mgoulao made their first contribution in https://github.com/Farama-Foundation/Gymnasium-Robotics/pull/35
  • @araffin made their first contribution in https://github.com/Farama-Foundation/Gymnasium-Robotics/pull/39
  • @RedTachyon made their first contribution in https://github.com/Farama-Foundation/Gymnasium-Robotics/pull/52
  • @Kallinteris-Andreas made their first contribution in https://github.com/Farama-Foundation/Gymnasium-Robotics/pull/54

- Python
Published by rodrigodelazcano about 3 years ago

gymnasium-robotics - v1.0.1: Deprecate package name (`gym_robotics`->`gymnasium_robotics`)

What's Changed

The PyPi package name for this repository will be changed in future releases and integration with Gymnasium. The new name will be gymnasium_robotics and installation will be done with pip install gymnasium_robotics instead of pip install gym_robotics.

The code for gym_robotics will be kept in the repository branch gym-robotics-legacy

Bug Fix

  • Remove the warning of duplicated registration of the environment MujocoHandBlockEnv @leonasting

- Python
Published by rodrigodelazcano over 3 years ago

gymnasium-robotics - v1.0.0: Update to Gym v0.26 and new mujoco bindings

This new release comes with the following changes:

  • Compatibility with gym v0.26. Previous gym versions won't be compatible with this release. @rodrigodelazcano
  • Added new environment versions that depend on the new mujoco python bindings. @rodrigodelazcano
  • Old environment versions that depend on mujoco_py are still kept but will be unmaintained moving forward. @rodrigodelazcano
  • New utility methods for GoalEnv class as suggested in #16 . compute_terminated and compute_truncated @rodrigodelazcano

The new versions of the environments that depend on mujoco bindings were validated with respect to the old versions of mujoco_py. The benchmark was performed using TQC + HER (sb3 implementation) with the same hyperparameters for both environment versions. The results can be seen here: https://wandb.ai/rodrigodelazcano/gym_robotics/reports/Benchmark-Gym-Robotics-SB3--VmlldzoyMjc3Mzkw

- Python
Published by rodrigodelazcano over 3 years ago

gymnasium-robotics - v0.1.0: Gym update

What's Changed

  • Change workflow name by @vwxyzjn in https://github.com/Farama-Foundation/gym-robotics/pull/4
  • Adopt gym>=0.22 reset signature by @vwxyzjn in https://github.com/Farama-Foundation/gym-robotics/pull/8
  • Use gym>=0.22 as the core dependency by @vwxyzjn in https://github.com/Farama-Foundation/gym-robotics/pull/9

Installation Demo

python pip install gym-robotics pip install mujoco_py wget https://mujoco.org/download/mujoco210-linux-x86_64.tar.gz mkdir -p ~/.mujoco tar -xzf mujoco210-linux-x86_64.tar.gz -C ~/.mujoco export LD_LIBRARY_PATH=$LD_LIBRARY_PATH:~/.mujoco/mujoco210/bin export LD_LIBRARY_PATH=$LD_LIBRARY_PATH:/usr/lib/nvidia asciicast

Full Changelog: https://github.com/Farama-Foundation/gym-robotics/compare/v0.0.2...v0.1.0

- Python
Published by vwxyzjn almost 4 years ago

gymnasium-robotics - v0.0.2

What's Changed

  • Migrate robotics environments from OpenAI Gym by @seungjaeryanlee in https://github.com/Farama-Foundation/gym-robotics/pull/1
  • Use Gym plugin system by @JesseFarebro in https://github.com/Farama-Foundation/gym-robotics/pull/2
  • Setup github actions to publish on PyPi by @vwxyzjn in https://github.com/Farama-Foundation/gym-robotics/pull/3

New Contributors

  • @seungjaeryanlee made their first contribution in https://github.com/Farama-Foundation/gym-robotics/pull/1
  • @JesseFarebro made their first contribution in https://github.com/Farama-Foundation/gym-robotics/pull/2
  • @vwxyzjn made their first contribution in https://github.com/Farama-Foundation/gym-robotics/pull/3

Full Changelog: https://github.com/Farama-Foundation/gym-robotics/commits/v0.0.2

- Python
Published by vwxyzjn about 4 years ago