Stanford · MIT · 11 MB
Stanford TidyBot
Stanford TidyBot developed by Jimmy Wu and collaborators from the Interactive Perception and Robot Learning Lab at Stanford University.
The first time takes a moment
Opening this model fetches it from the Menagerie repository — the scene, the robot and the meshes it references, about 11 MB — and keeps it here, so anyone opening it after you gets it straight away. Nothing of yours is uploaded in the process: the simulation itself runs in your tab, on MuJoCo compiled to WebAssembly.
What you can do with it here
- Simulate it under gravity and contact, write a controller in JavaScript, shove it with the mouse, and run the four repeatable tests that answer with a number rather than an impression.
- Look inside it: the kinematic tree, every joint and its range, collision geometry against the visual mesh, where the mass actually sits.
- Convert it to URDF for ROS, or to USD for Isaac Sim — each direction states what it could not carry across.
- Author a movement for it and check the motors it declares can actually perform what you drew.
Whose model this is
The Stanford TidyBot model is MIT licensed and was published by Stanford as part of the MuJoCo Menagerie, curated by Google DeepMind. This page is a way to open it, not a claim on it — the folder upstream carries the licence text, the changelog and the README its authors wrote:
https://github.com/google-deepmind/mujoco_menagerie/tree/main/stanford_tidybot
If you use the model in your own work, credit that folder rather than this page. If it has changed upstream since we fetched it, the copy here is the one you get — say so in a bug report and it will be re-fetched.
Nearby in the collection
- Google Robot — Google, Apache 2.0
- Hello Robot Stretch 2 — Hello Robot, BSD 3-Clause
- Hello Robot Stretch 3 — Hello Robot, Apache 2.0
- Pal Tiago — PAL Robotics, Apache 2.0
- Pal Tiago Dual — PAL Robotics, Apache 2.0
- Robot soccer kit omnidirectional — Robot, MIT