google-deepmind / google-deepmind/mujoco_menagerie

Add Petoi Bittle X

Open
#286 0 comments 0 reactions 0 assignees View on GitHub
new-model
Dominant language
Python
Stars
4k
Forks
558
Avg merge
5d 9h
Merged PRs (30d)
4

Description

If you would like to see a new model added to the Menagerie collection, please provide us with the following information _before_ sending a pull request. (Ideally, before you start working on converting or tuning the model for MuJoCo.)

I'd like to contribute a Petoi Bittle X V1 model to Menagerie. The MJCF is ready and tested - just checking if you'd accept a PR before I adapt it to your format.

**URL for the existing model source (e.g. URDF)**
Original URDF: [PetoiCamp/ros_opencat](https://github.com/PetoiCamp/ros_opencat) (MIT License)
My MJCF conversion: [MarcHesse/bittle-mujoco](https://github.com/MarcHesse/bittle-mujoco)

**License terms of the upstream model**
MIT (meshes and URDF). My MJCF conversion is Apache 2.0.

**What do you think is required to convert and tune this model for MuJoCo?**
Already done. The model has 8 position actuators (2 per leg), 40 OBJ meshes with collision geometries, IMU sensor site, foot contact sensors, verified stand keyframe, and mass properties based on manufacturer specs (~265g with plastic servos). I have the physical robot for parameter tuning. The Bittle X is a small open-source quadruped (ESP32, 8 DOF) - it would be the first consumer-priced quadruped in the collection.

**Are you proposing to help develop and maintain the new model?**
Yes. I use this model daily for locomotion research ([MH-FLOCKE](https://github.com/MarcHesse/mhflocke)).

Contributor guide

Open the contributing guide

Research direction

Review the existing Menagerie collection conventions, then inspect the linked PetoiCamp/ros_opencat URDF and MarcHesse/bittle-mujoco MJCF. Confirm that the proposed model meets the repository's required format, licensing expectations, and validation standards; done means a maintainable pull request can be submitted for review.

Written by the indexing model from the issue text.

Assessment

Domain
robotics
Issue type
Feature
Difficulty
4/5
Estimated time
3-5 days
Activity status
Quiet
Clarity
Mostly clear
Newbie friendliness
48/100

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.