dora-mujoco-husky

A MuJoCo-based Clearpath Husky simulation node for the Dora framework. This node provides a physics-accurate simulation of the Husky robot that can be controlled via velocity commands.

Features

• Real-time physics simulation using MuJoCo
• Accurate Husky robot model with proper inertial properties
• Velocity-based control interface
• Position and velocity feedback
• Gamepad control support

Getting Started

Prerequisites

• Python 3.8 or higher
• MuJoCo 3.1.6 or higher

Installation

1. Create and activate a Python virtual environment:

uv venv -p 3.11 --seed
source .venv/bin/activate

2. Install the package:

uv pip install -e .

Running the Demo

1. Make sure you have a gamepad connected
2. Build the demo:

dora build demo.yml --uv

3. Run the demo:

dora run demo.yml --uv

Usage

The node accepts velocity commands and outputs position/velocity data:

Inputs

• cmd_vel: Velocity command array with the format:
  • Index 0: Linear velocity (m/s), range: [-2.0, 2.0]
  • Index 5: Angular velocity (rad/s), range: [-3.0, 3.0]

Outputs

• position: Robot position [x, y, z] in world coordinates
• velocity: Robot velocity [vx, vy, vz] in world coordinates

Example Configuration

nodes:
  - id: mujoco_husky
    build: pip install -e .
    path: dora-mujoco-husky
    inputs:
      cmd_vel: input_source/cmd_vel
    outputs:
      - position
      - velocity

Development

Code Formatting

Format code using ruff:

uv pip install ruff
uv run ruff check . --fix

Linting

Run linting checks:

uv run ruff check .

Testing

Run tests using pytest:

uv pip install pytest
uv run pytest .

License

This project is licensed under the MIT License - see the LICENSE file for details.

Acknowledgments

• Based on the Clearpath Robotics Husky robot
• Uses the MuJoCo physics engine
• Built with the Dora framework