gymnasium-robotics

A collection of robotics simulation environments for reinforcement learning

https://github.com/farama-foundation/gymnasium-robotics

Science Score: 36.0%

This score indicates how likely this project is to be science-related based on various indicators:

○
CITATION.cff file
✓
codemeta.json file
Found codemeta.json file
✓
.zenodo.json file
Found .zenodo.json file
○
DOI references
○
Academic publication links
✓
Committers with academic emails
3 of 32 committers (9.4%) from academic institutions
○
Institutional organization owner
○
JOSS paper metadata
○
Scientific vocabulary similarity
Low similarity (14.3%) to scientific vocabulary

Keywords

d4rl gymnasium mujoco reinforcement-learning robotics simulation

Keywords from Contributors

gym gym-environment gridworld-environment offline-rl multiagent-reinforcement-learning multi-agent-reinforcement-learning deepmind-lab deepmind-control-suite autonomous-driving baselines

Last synced: 6 months ago · JSON representation

Repository

A collection of robotics simulation environments for reinforcement learning

Basic Info

Host: GitHub
Owner: Farama-Foundation
License: mit
Language: Python
Default Branch: main
Homepage: https://robotics.farama.org/
Size: 373 MB

Statistics

Stars: 752
Watchers: 15
Forks: 117
Open Issues: 20
Releases: 14

Topics

d4rl gymnasium mujoco reinforcement-learning robotics simulation

Created over 4 years ago · Last pushed 6 months ago

Metadata Files

Readme Funding License Code of conduct Citation

This library contains a collection of Reinforcement Learning robotic environments that use the Gymnasium API. The environments run with the MuJoCo physics engine and the maintained mujoco python bindings.

The documentation website is at robotics.farama.org, and we have a public discord server (which we also use to coordinate development work) that you can join here: https://discord.gg/YymmHrvS

Installation

To install the Gymnasium-Robotics environments use pip install gymnasium-robotics

These environments also require the MuJoCo engine from Deepmind to be installed. Instructions to install the physics engine can be found at the MuJoCo website and the MuJoCo Github repository.

Note that the latest environment versions use the latest mujoco python bindings maintained by the MuJoCo team. If you wish to use the old versions of the environments that depend on mujoco-py, please install this library with pip install gymnasium-robotics[mujoco-py]

We support and test for Linux and macOS. We will accept PRs related to Windows, but do not officially support it.

Environments

Gymnasium-Robotics includes the following groups of environments:

Fetch - A collection of environments with a 7-DoF robot arm that has to perform manipulation tasks such as Reach, Push, Slide or Pick and Place.
Shadow Dexterous Hand - A collection of environments with a 24-DoF anthropomorphic robotic hand that has to perform object manipulation tasks with a cube, egg-object, or pen. There are variations of these environments that also include data from 92 touch sensors in the observation space.
MaMuJoCo - A collection of multi agent factorizations of the Gymnasium/MuJoCo environments and a framework for factorizing robotic environments, uses the pettingzoo.ParallelEnv API.

The D4RL environments are now available. These environments have been refactored and may not have the same action/observation spaces as the original, please read their documentation:

Maze Environments - An agent has to navigate through a maze to reach certain goal position. Two different agents can be used: a 2-DoF force-controlled ball, or the classic Ant agent from the Gymnasium MuJoCo environments. The environment can be initialized with a variety of maze shapes with increasing levels of difficulty.
Adroit Arm - A collection of environments that use the Shadow Dexterous Hand with additional degrees of freedom for the arm movement. The different tasks involve hammering a nail, opening a door, twirling a pen, or picking up and moving a ball.
Franka Kitchen - Multitask environment in which a 9-DoF Franka robot is placed in a kitchen containing several common household items. The goal of each task is to interact with the items in order to reach a desired goal configuration.

WIP: generate new D4RL environment datasets with Minari.

Multi-goal API

The robotic environments use an extension of the core Gymnasium API by inheriting from GoalEnv class. The new API forces the environments to have a dictionary observation space that contains 3 keys:

observation - The actual observation of the environment
desired_goal - The goal that the agent has to achieved
achieved_goal - The goal that the agent has currently achieved instead. The objective of the environments is for this value to be close to desired_goal

This API also exposes the function of the reward, as well as the terminated and truncated signals to re-compute their values with different goals. This functionality is useful for algorithms that use Hindsight Experience Replay (HER).

The following example demonstrates how the exposed reward, terminated, and truncated functions can be used to re-compute the values with substituted goals. The info dictionary can be used to store additional information that may be necessary to re-compute the reward, but that is independent of the goal, e.g. state derived from the simulation.

```python import gymnasium as gym

env = gym.make("FetchReach-v3") env.reset() obs, reward, terminated, truncated, info = env.step(env.action_space.sample())

The following always has to hold:

assert reward == env.computereward(obs["achievedgoal"], obs["desiredgoal"], info) assert truncated == env.computetruncated(obs["achievedgoal"], obs["desiredgoal"], info) assert terminated == env.computeterminated(obs["achievedgoal"], obs["desired_goal"], info)

However goals can also be substituted:

substitutegoal = obs["achievedgoal"].copy() substitutereward = env.computereward(obs["achievedgoal"], substitutegoal, info) substituteterminated = env.computeterminated(obs["achievedgoal"], substitutegoal, info) substitutetruncated = env.computetruncated(obs["achievedgoal"], substitutegoal, info) ```

The GoalEnv class can also be used for custom environments.

Project Maintainers

Main Contributors: Rodrigo Perez-Vicente, Kallinteris Andreas, Jet Tai

Maintenance for this project is also contributed by the broader Farama team: farama.org/team.

Citation

If you use this in your research, please cite: @software{gymnasium_robotics2023github, author = {Rodrigo de Lazcano and Kallinteris Andreas and Jun Jet Tai and Seungjae Ryan Lee and Jordan Terry}, title = {Gymnasium Robotics}, url = {http://github.com/Farama-Foundation/Gymnasium-Robotics}, version = {1.4.0}, year = {2024}, }

Owner

Name: Farama Foundation
Login: Farama-Foundation
Kind: organization
Email: contact@farama.org

Website: farama.org
Twitter: FaramaFound
Repositories: 49
Profile: https://github.com/Farama-Foundation

The Farama foundation is a nonprofit organization working to develop and maintain open source reinforcement learning tools.

GitHub Events

Total

Create event: 12
Release event: 2
Issues event: 21
Watch event: 208
Delete event: 10
Issue comment event: 62
Push event: 56
Pull request review comment event: 15
Pull request review event: 18
Pull request event: 33
Fork event: 33

Last Year

Create event: 12
Release event: 2
Issues event: 21
Watch event: 208
Delete event: 10
Issue comment event: 62
Push event: 56
Pull request review comment event: 15
Pull request review event: 18
Pull request event: 33
Fork event: 33

Committers

Last synced: over 1 year ago

All Time

Total Commits: 295
Total Committers: 32
Avg Commits per committer: 9.219
Development Distribution Score (DDS): 0.783

Past Year

Commits: 41
Committers: 7
Avg Commits per committer: 5.857
Development Distribution Score (DDS): 0.512

Top Committers

Name	Email	Commits
Rodrigo de Lazcano	r**6@g**m	64
Kallinteris Andreas	k**s@p**m	43
Kallinteris Andreas	3****s	36
Rodrigo de Lazcano Perez-Vicente	r**o@u**u	27
snow-fox	t**t@h**m	26
Seungjae Ryan Lee	s**e@g**m	22
Manuel Goulão	m**o@g**m	13
Jet	3****s	12
Feng Gu	f**0@g**m	7
Jordan Terry	j**0@g**m	7
Martin Schuck	m**k@t**e	7
Costa Huang	c**g@o**m	5
J K Terry	j**y@g**m	3
Feng	5****u	3
Mark Towers	m**s@g**m	2
Fangyuan	f**6@g**m	2
Timo Friedl	m**l@t**m	1
Seth Pate	s**e@i**m	1
Aziz	4****e	1
Daniel CH Tan	d****7	1
Edward Hu	1****s	1
Jan Dohmen	3****n	1
Quentin GALLOUÉDEC	g**n@g**m	1
Jesse Farebrother	j**o@g**m	1
Gautham Vasan	g**9@g**m	1
Frank Röder	f**g@p**e	1
David Leins	d**s@t**e	1
Ariel Kwiatkowski	a**i@g**m	1
Antonin RAFFIN	a**n@e**g	1
Andrew Tan	a**5@g**m	1
and 2 more...

Committer Domains (Top 20 + Academic)

ensta.org: 1 techfak.uni-bielefeld.de: 1 proton.me: 1 timofriedl.com: 1 tum.de: 1 umd.edu: 1

Issues and Pull Requests

Last synced: 6 months ago

All Time

Total issues: 69
Total pull requests: 164
Average time to close issues: about 2 months
Average time to close pull requests: 7 days
Total issue authors: 57
Total pull request authors: 28
Average comments per issue: 3.23
Average comments per pull request: 0.82
Merged pull requests: 143
Bot issues: 0
Bot pull requests: 3

Past Year

Issues: 19
Pull requests: 36
Average time to close issues: 15 days
Average time to close pull requests: 5 days
Issue authors: 19
Pull request authors: 9
Average comments per issue: 2.05
Average comments per pull request: 0.22
Merged pull requests: 26
Bot issues: 0
Bot pull requests: 3

View more stats

Top Authors

Issue Authors

Kallinteris-Andreas (5)
llewynS (3)
matinmoezzi (2)
aalmuzairee (2)
sebidoe (2)
ekkooee7 (1)
younik (1)
AlexandreBrown (1)
amacati (1)
greg3566 (1)
R-Liebert (1)
ozhanozen (1)
wxforvip (1)
RayYoh (1)
Rancho-zhao (1)

Pull Request Authors

Kallinteris-Andreas (85)
rodrigodelazcano (47)
jjshoots (12)
mgoulao (10)
pseudo-rnd-thoughts (4)
dependabot[bot] (3)
xihuai18 (2)
Josh00-Lu (2)
mishmish66 (2)
siddarth-c (2)
timofriedl (2)
RussTedrake (2)
HridayM25 (2)
Tanmay692004 (2)
nicehiro (2)

Top Labels

Issue Labels

good first issue (3) enhancement (1) help wanted (1)

Pull Request Labels

dependencies (3) github_actions (3)

Packages

Total packages: 3
Total downloads:
- pypi 30,527 last-month
Total docker downloads: 155

Total dependent packages: 6
(may contain duplicates)
Total dependent repositories: 13
(may contain duplicates)
Total versions: 36
Total maintainers: 3

pypi.org: gymnasium-robotics

RL Robotics environments with Gymnasium API.

Homepage: https://farama.org
Documentation: https://robotics.farama.org
License: MIT License
Latest release: 1.4.1
published 7 months ago

Versions: 10
Dependent Packages: 6
Dependent Repositories: 13
Downloads: 30,527 Last month
Docker Downloads: 155

Rankings

Dependent packages count: 2.4%

Docker downloads count: 2.7%

Stargazers count: 3.6%

Average: 3.8%

Dependent repos count: 4.0%

Downloads: 4.7%

Forks count: 5.6%

Maintainers (3)

jkterry pseudornd rodrigodelazcano

Last synced: 6 months ago

proxy.golang.org: github.com/farama-foundation/gymnasium-robotics

Documentation: https://pkg.go.dev/github.com/farama-foundation/gymnasium-robotics#section-documentation
License: mit
Latest release: v1.3.1
published over 1 year ago

Versions: 12
Dependent Packages: 0
Dependent Repositories: 0

Rankings

Dependent packages count: 6.3%

Average: 6.5%

Dependent repos count: 6.7%

Last synced: about 1 year ago

proxy.golang.org: github.com/Farama-Foundation/Gymnasium-Robotics

Documentation: https://pkg.go.dev/github.com/Farama-Foundation/Gymnasium-Robotics#section-documentation
License: mit
Latest release: v1.4.1
published 7 months ago

Versions: 14
Dependent Packages: 0
Dependent Repositories: 0

Rankings

Dependent packages count: 6.5%

Average: 6.7%

Dependent repos count: 6.9%

Last synced: 6 months ago

Dependencies

requirements.txt pypi

cloudpickle >=1.2.0
gym >=0.22
mujoco_py >=1.50,<2.0
numpy >=1.18.0

setup.py pypi

cloudpickle >=1.2.0
gym >=0.22
importlib_metadata >=4.8.1
numpy >=1.18.0

test_requirements.txt pypi

pytest * test
pytest-forked * test

gymnasium-robotics

Science Score: 36.0%

Keywords

Keywords from Contributors

Repository

Basic Info

Statistics

Topics

Metadata Files

README.md

Installation

Environments

Multi-goal API

The following always has to hold:

However goals can also be substituted:

Project Maintainers

Citation

Owner

GitHub Events

Total

Last Year

Committers

All Time

Past Year

Top Committers

Committer Domains (Top 20 + Academic)

Issues and Pull Requests

All Time

Past Year

Top Authors

Issue Authors

Pull Request Authors

Top Labels

Issue Labels

Pull Request Labels

Packages

pypi.org: gymnasium-robotics

Rankings

Maintainers (3)

proxy.golang.org: github.com/farama-foundation/gymnasium-robotics

Rankings

proxy.golang.org: github.com/Farama-Foundation/Gymnasium-Robotics

Rankings

Dependencies