KO | EN

GitLite

Minimalist Open Source Repository Search Engine

gitlite — search

#dsh #dsh-plugin #ai-agent #developer-tools #codex #typescript #cli #deepseek-harness #claude-code #ai-agents #mcp #rust

mujoco-robotics-usage

★ 54 Open GitHub ↗

Easy-to-follow, Intuitive MuJoCo tutorial codes for robotics algorithms.

// repository documentation

Was this content helpful?

★ 0 (0 ratings)

# mujoco-robotics-usage
This repo provides minimal hands-on code for MuJoCo Robotics Algorithms.

(This repository only covers the MuJoCo simulation. Realworld is not included)

MuJoCo related code is employed from following repos:

* [MuJoCo] https://github.com/deepmind/mujoco
* [YAMT] https://github.com/sjchoi86/yet-another-mujoco-tutorial-v3

Some of the robot models referenced mujoco_menagerie
* [mujoco_menagerie] https://github.com/deepmind/mujoco_menagerie

## Prerequisites

This repo is tested on following environment:

* Ubuntu: 20.04
* Python: 3.8.10
* mujoco: 2.3.2

### Install dependencies

Mujoco Engine
```bash
pip install mujoco

pip install mujoco-python-viewer
```

### Descriptions
Below is a list of files and their descriptions:

* Kinematic Solver
 1. Solve inverse kinematics in various method with 
 * [General]
 
 <img src="./readme/gifs/ik_general.gif" width="49%" />
 <img src="./readme/gifs/wholebody_ik.gif" width="49%" />

* [Augmented],
        * [Nullspace projection]
        * [Repelling]
        * [RRT*]
        
        
* Trajectory Planning method
    1. [Task space planning]
        * [Quintic]

* [Minimum Jerk]
 * [Linear movement]
 2. [Velocity profile method]: 
 * [Trapezoidal]
 * [s-Spline method]
 
* Demos
 1. [Pick-n-Place]
 
 <img src="./readme/gifs/manipulator_pnp.gif" width="100%" alt="Pick-n-place multiple objects in succession on table-top scene.">
 
 
* Mobile Planning method
 1. [Mobile Velocity Control]
 
 <img src="./readme/gifs/mobile_planner.gif" width="100%" alt="Husky with UR just moving forward.">
 
 2. [SLAM + RRT* + Navigation]
 * 2D LiDAR SLAM (occupancy grid mapping with 72 rangefinder sensors)
 * RRT* global path planning on SLAM map
 * Local planners: Pure Pursuit, Stanley Controller, MPPI
 * Dynamic obstacle avoidance with local path deformation

Code: `code/mobile/demo_husky_06_slam.ipynb` ~ `demo_husky_10_rrt_w_avoidance.ipynb`
 
 
* Point-cloud
 1. [Point-cloud Projection]
 
 <img src="./readme/image/project_pcd.png" width="100%" alt="Project point-cloud onto table from ego-centric view.">
 
 
 <img src="./readme/gifs/pcd_projection.gif" width="100%" alt="Project point-cloud onto table from ego-centric view.">

2. [Colored pointcloud]
 
 <img src="./readme/image/input_rgbd.png" width="49%" alt="RGBD and colored pointcloud visualization - Two-Tiered Bookshelf"/>
 <img src="./readme/image/colored_pcd.png" width="49%" alt="RGBD and colored pointcloud visualization - Three-Tiered Bookshelf"/>
 
 When input RGBD has given, converts it to corresponding colored pointcloud in a row.

Code: 
        [demo_colored_pcd.ipynb](code/point-cloud/demo_colored_pcd.ipynb)

3. [Iterative Closet Point]
    4. [Extrinsic calibration]
    
    
* Segmentations
    * [Unseen Object Clustering (UCN)](https://github.com/NVlabs/UnseenObjectClustering)
    * [Segment-Anything (SAM)](https://github.com/facebookresearch/segment-anything)

* Miscellaneous
 * [Multi-vivwer] Add Multi-Viewer toy examples: Get images from camera defined in an XML(MJCF) files.
 
 <img src="./readme/gifs/multi-viewer-1.gif" width="49%" />
 <img src="./readme/gifs/multi-viewer-2.gif" width="49%" />>

mujoco-robotics-usage

Explore Similar Repositories

Was this content helpful?

Recent Feedback

Download README