robosuite
SoftwareMuJoCo-based robot-learning environments with configurable robots, grippers, controllers and observations; the reviewed v1.5.2 setup requires matching rendering and action conventions. [1]
Installation & first result →Official websiteGetting started
- For
- Robot-learning researchers building controlled simulation experiments [1]
- Source rights
- Some licenses are unresolved. Review licenses
- Unknowns
- compute requirements not fully established. Review gaps
Source researched · Not tested by us
Sources checked · Research details
Licenses & related hardware
- License
- MIT · license not fully assessed for documentation · Details by part
Overview
robosuite provides MuJoCo-based simulation environments for robot learning. Its modular interface combines robots, grippers, controllers, arenas and objects into tasks, with particular emphasis on manipulation. The reviewed source is v1.5.2; the v1.5 family adds broader embodiments and composite controllers. These capabilities describe the framework, not a demonstrated policy result on physical hardware. [1] [2] [7]
Is this for you?
- Workflows
- Configure robots, grippers and controllers for a manipulation task
- Choose low-dimensional or camera observations
- Collect or replay demonstrations
- Implement and register custom composite controllers
- Known limitations
- Camera observations require off-screen rendering in the documented setup
- Default camera image convention is OpenGL flipped
- Task success does not itself end an episode; documented rollouts use a fixed horizon
- Changing the controller changes the action space
What you need
- Operating system
- Linux
- macOS
Official support statement; Windows troubleshooting is separately documented, not equivalent support.
Installation & first result
Open the upstream getting-started guide [5] · Project documentation [4] [5]
- Installation
Use an isolated Python environment; the guide suggests Python 3.10. Install the selected release and run the documented random-action demo; macOS default viewer uses mjpython.
Editorial sequence; no installation or demo executed.
[5] · Editorial assessment
Hardware, models & data
Project notes
Choose a release and a rendering route
The installation guide officially supports macOS and Linux and recommends an isolated Python environment, with Python 3.10 in its Conda example. The selected package metadata requires MuJoCo 3.3.0 or newer, pins mink to 0.0.5 and lists additional numerical and rendering dependencies. Its broad Python metadata alone does not establish that every Python 3 version works with the full dependency set. Record the actual resolved versions for an experiment. [5] [6]
The documented first check is the random-action demo. On macOS, the default MuJoCo viewer uses mjpython rather than the ordinary Python launcher. The same guide includes Windows troubleshooting, but its official support statement names macOS and Linux. Optional devices and integrations need additional dependencies. This review did not install the package or run its demo. [5]
Observations and actions are configuration choices
The environment guide constructs a TwoArmLift task with Sawyer and Panda robots and a composite controller loaded through load_composite_controller_config. Robots, grippers, robot arrangement and controller configuration can all change. A controller change can also change the action space; an action vector from one configuration should not be assumed to mean the same thing in another. [7]
For a first integration, use the environment's action_spec bounds and the documented reset/step loop. Observations are dictionaries rather than one universal flat vector. They can contain proprioception, object state, RGB images and depth, depending on the enabled modalities. Record those choices with the policy configuration so a comparison uses the same information at training and evaluation. This recording recommendation is an editorial reproducibility step. [7]
The image-based example enables off-screen rendering and camera observations while disabling the on-screen renderer. A headless process therefore still needs the rendering route appropriate to its selected observations. The default image convention is OpenGL-style, described upstream as flipped; IMAGE_CONVENTION can switch to OpenCV-style output. That difference matters when reusing camera preprocessing. [7]
Success and termination are different signals
The guide distinguishes sparse task-success rewards from shaped rewards. It also states that reaching the success criterion does not terminate the episode: environments continue to the configured horizon. A comparison that treats the returned termination signal as a success flag would therefore measure a different quantity. Preserve the reward setting, success definition, control frequency and horizon with evaluation results. No success rate or timing measurement was produced here. [7]
Custom composite controllers require more than a configuration name. The extension guide subclasses a composite controller, registers it, implements its behavior, imports the class and supplies matching configuration. Its third-party example illustrates the route; it is not evidence that an arbitrary external controller is compatible. [8]
The root license is MIT and separately identifies included MuJoCo portions under Apache-2.0. It does not establish blanket rights to every external model, renderer or dataset a user might select. This entry covers the simulation framework and its random-action setup route, without downloading a policy or demonstration dataset or claiming hardware transfer. [3] [5]
What we checked
Source research only; no installation, physical build or performance test performed. No affiliate links added.
Not yet established
- Requirements: host computer
- A step-by-step setup summary
- Version-specific hardware compatibility (ecosystem tags alone do not establish it)
Updates
Source files and licenses
Available project sources support research coverage; complete build and commercial qualification remain separate.
| Layer | Availability / license |
|---|---|
| software | available · MIT Root license separately identifies included MuJoCo portions under Apache-2.0; external dependencies and assets retain their own terms. [3] |
| documentation | available · License not fully assessed Source documentation available; separate documentation-wide rights not established. [4] [5] |
This profile selects the environment framework and random-action setup route, not a pretrained policy or demonstration dataset.
References
Each source links to the exact revision we read. “Used for” shows where this page relies on it.
Project sources
- S001 ARISE-Initiative/robosuite: README.md
Revision, date & where cited
github.com · revision
824ac14· Accessed Oct 1, 2026
Used for: Overview, Is this for you? - S002 ARISE-Initiative/robosuite: commit 824ac14
- S003 ARISE-Initiative/robosuite: LICENSE
Revision, date & where cited
github.com · revision
824ac14· Accessed Oct 1, 2026
Used for: Success and termination are different signals, Source files and licenses - S004 ARISE-Initiative/robosuite: files
Revision, date & where cited
github.com · revision
824ac14· archive file inventory · Accessed Oct 1, 2026
Used for: Installation & first result, Source files and licenses - S005 robosuite v1.5.2: docs/installation.md
Revision, date & where cited
github.com · revision
824ac14· Accessed Oct 1, 2026
Used for: What you need, Installation & first result, Choose a release and a rendering route, Success and termination are different signals and 1 more - S006 robosuite v1.5.2: setup.py
Revision, date & where cited
github.com · revision
824ac14· Accessed Oct 1, 2026
Used for: What you need, Choose a release and a rendering route - S007 robosuite v1.5.2: docs/modules/environments.md
Revision, date & where cited
github.com · revision
824ac14· Accessed Oct 1, 2026
Used for: Overview, Is this for you?, Hardware, models & data, Observations and actions are configuration choices and 1 more - S008 robosuite v1.5.2: docs/tutorials/add_controller.md
Revision, date & where cited
github.com · revision
824ac14· Accessed Oct 1, 2026
Used for: Is this for you?, Hardware, models & data, Success and termination are different signals
Videos
- S009 robosuite v1.3 release (YouTube video)
Video uploaded by Yuke Zhu, linked from the project’s own documentation. Not reviewed by us for accuracy.Revision, date & where cited
YouTube video Yuke Zhu · youtube.com · Accessed Oct 4, 2026
Correction guidance · Discussion is not enabled · Profile format 0.6.0-draft.1
Change history
2026-10-04 — Added 1 YouTube reference(s) (S009): official channel or videos linked from the pinned upstream sources. Existing claims, source-check dates and evidence are unchanged.