Unitree-Go2 Robot: SLAM and Nav2 Demos

By: Hossein Naderi

Welcome to the Unitree-Go2 Robot SLAM and Nav2 repository! This project demonstrates the integration of SLAM (Simultaneous Localization and Mapping) and autonomous navigation using Nav2 on the Unitree-Go2 robot. The repository includes demonstration videos, source code, and comprehensive documentation to assist in replicating and building upon the showcased capabilities.​:contentReference[oaicite:0]{index=0}

Table of Contents

1. Overview
2. Demo Videos
3. Project Structure
4. Workspace Setup
5. Launch Files
6. Detailed Project Explanations

Overview

• Robot Platform: Unitree-Go2
• SLAM: ROS 2 RTAB-Map
• Navigation: ROS 2 Nav2

Key Features

• Real-time mapping in indoor and outdoor environments
• Obstacle avoidance using Nav2 and sensor data
• Sensor fusion combining RGB Depth Camera with LiDAR point cloud
• Autonomous path planning to specified goals
• Facial recognition capabilities for identifying individuals during navigation

Demo Videos

1. SLAM Demo Video
   Demonstration of integrated SLAM with Unitree-Go2 using RTAB-Map in ROS2.

   

2. Autonomous Navigation Demo Video
   Demonstration of integrated Autonomous Navigation with Unitree-Go2 using Nav2 Stack in ROS2.

   

Dependencies

• Unitree SDK: https://github.com/unitreerobotics/unitree_ros2
• LiDAR SDK: https://github.com/RoboSense-LiDAR/rslidar_sdk
• LiDAR Messages: https://github.com/RoboSense-LiDAR/rslidar_msg

Workspace Setup

1. Create your workspace:

   mkdir -p ~/ws/src
   cd ~/ws/src
   

1. Clone this repository:

   git clone https://github.com/h-naderi/unitree-go2-slam-nav2.git

   

1. Navigate to the repository:

   cd unitree-go2-slam-nav2
   

1. Import dependencies using vcs:

   vcs import < dependencies.repo
   

1. Install dependencies and build the workspace:

   cd ~/ws
   rosdep install --from-paths src --ignore-src -r -y
   colcon build
   

1. Source the workspace:

   source ~/ws/install/setup.bash
   

Launch Files

1. Launch the Intel RealSense camera node:

    ros2 launch realsense2_camera rs_launch.py enable_gyro:=true enable_accel:=true unite_imu_method:=1 enable_infra1:=true enable_infra2:=true enable_sync:=true
    

1. Launch everything on the robot for mapping with the LiDAR only:

    ros2 launch go2_slam_nav mapping.launch.py
    

1. Launch the robot with the RealSense camera and the LiDAR:

    ros2 launch go2_slam_nav mapping_camera.launch.py
    

1. Launch the navigation:

    ros2 launch go2_slam_nav nav.launch.py
    

1. Launch the exploration node only:

    ros2 launch frontier frontier_update.launch.xml
    

Feel free to reach out if you have any questions or need further assistance with setting up the project.

Detailed Project Explanations

For comprehensive explanations and insights into each project component, please refer to my portfolio:

• SLAM Project: SLAM Project
• Autonomous Navigation: Navigation Project