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

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]

       

     • [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]

     

• Mobile Planning method

  1. [Mobile Velocity Control]

     

  2. [Global planner]
     • [A*]
     • [RRT]
     • [RRT*]
  3. [Local planner]
     • [Pure-pursuit]

• Point-cloud

  1. [Point-cloud Projection]

     

     

  2. [Colored pointcloud]

     

     When input RGBD has given, converts it to corresponding colored pointcloud in a row.

     Code: demo_colored_pcd.ipynb

  3. [Iterative Closet Point]

  4. [Extrinsic calibration]

• Segmentations

  • Unseen Object Clustering (UCN)
  • Segment-Anything (SAM)

• Miscellaneous

  • [Multi-vivwer] Add Multi-Viewer toy examples: Get images from camera defined in an XML(MJCF) files.

    >