MRAC-robotic-spatial-analysis

Overview

This is the repository for MRAC 24/25 Workshop 2.2, which focuses on advanced 3D scanning and spatial analysis of building environments using a combination of photogrammetry, 3D LiDAR, robot platform, robot dog, and drone. This project emphasizes the use of robotic systems to autonomously or via teleoperation gather and analyze spatial data in different environments. Each system contributes to separate parts of the scanning and analysis process, providing a comprehensive approach to spatial data collection and 3D modeling.

Table of Contents

1. Overview
2. Key Technologies
3. Project Structure
4. Installation
   • Required Operating Systems
   • Foxglove Studio Installation
   • SSH Server Setup
   • PCL Tools
5. Usage
   • Husky Connection
   • Launching ROS2 Nodes
   • Foxglove Bridge Setup

Key Technologies

• Photogrammetry: High-resolution 3D model generation from photographs.
• 3D LiDAR: Laser-based scanning for precise spatial mapping.
• Robot Platforms (Husky A200): Ground-based robotic platforms equipped for 3D scanning and data collection.
• Robot Dog(unitree GO2): Agile robots designed for scanning in challenging environments.
• Drone: Aerial data collection to complement ground-based scans.
• ROS (Robot Operating System): Used for communication and control of the rover and robot dog.

Project Structure

• Rover and Robot Dog: Teleoperated robotic platforms, using ROS for communication and control, designed for navigating and scanning different environments.
• Photogrammetry & LiDAR Systems: These technologies operate independently to gather detailed spatial data from different areas.
• Data Analysis: Spatial data is analyzed separately for each scanning method, enabling a comprehensive understanding of the environment.

Installation

Required Operating Systems: Ubuntu 22.04, Windows10/11

Foxglove Studio Installation

sudo snap install foxglove-studio

SSH Server

sudo apt install openssh-server

PCL Tools

sudo apt install pcl-tools

Usage

Husky Connection

1. Connect to the Husky robot:

   • Connect to the iaac_husky hotspot (WIFI password: EnterIaac22@).
   • Open a terminal on your computer and SSH into the Husky robot (user password: iaac):

     ssh [email protected]

2. Using tmux for terminal management:

   • Start a new tmux session:

     tmux new -s husky

     This command will start a new tmux session named husky.

3. Split the tmux terminal:

   • Once inside tmux, you can split the terminal window to launch multiple processes concurrently. To split the tmux terminal:
     • Press Ctrl + B, then release both keys and press % to split the terminal vertically.
     • To split horizontally, press Ctrl + B, then release both keys and press " (double quote).

4. Launch the ROS2 nodes in separate tmux panes:

   Terminal 1 (Mobile Base):

   • In the first tmux pane, launch the mobile base:

     ros2 launch /etc/clearpath/platform/launch/platform-service.launch.py

   Terminal 2 (LiDAR):

   • In the second tmux pane, launch the LiDAR:

     ros2 launch livox_ros_driver2 msg_MID360_launch.py

   Terminal 3 (Mapping):

   • In the third tmux pane, launch the mapping node:

     ros2 launch fast_lio mapping.launch.py

   Terminal 4 (Foxglove Bridge):

   • In the fourth tmux pane, launch the Foxglove bridge:

     ros2 launch foxglove_bridge foxglove_bridge_launch.xml

5. Open Foxglove Studio:

   • Open Foxglove Studio and set the web address to 10.42.0.1.
   • Load the panel from this repository to visualize the data.

Unitree Go2 Robot

1. Clone the repository:

   • Clone the MRAC-robot-spatial-analysis repository to your system.

2. Build the Docker image:

   • Note: You only need to build the image when using the Unitree Go2 robot.
   • Navigate to the go2_robot package directory and build the Docker image:

     cd MRAC-robot-spatial-analysis/go2_robot
     .docker/build_image.sh

3. Run the Docker image:

   • To run the image, use the following command:

     .docker/run_user.sh

   • Or, if you're using an Nvidia graphic card, run the image with the Nvidia runtime:

     .docker/run_user_nvidia.sh

4. Change the folder ownership:

   • To avoid permission issues, change the ownership of the workspace folder:

     sudo chown -R YOUR_USER_NAME /dev_ws

5. Terminal Setup for Go2 Robot:

   Terminal 1 (Go2 Bringup):

   • In the first terminal, bring up the robot dog:

     ros2 launch go2_bringup go2.launch.py

   Terminal 2 (RQT Visualization):

   • In the second terminal, use RQT to visualize the camera image or other sensor data:

     rqt

     You can use rqt_image_view to see the camera feed or other visualization plugins for other sensors.

   Terminal 3 (Teleoperation):

   • In the third terminal, use the keyboard teleoperation for controlling the robot:

     ros2 run teleop_twist_keyboard teleop_twist_keyboard

     This will allow you to control the Unitree Go2 robot using the keyboard.

   Terminal 4 (Map Saver Service):

   • In the fourth terminal, use the map saver service to save the generated voxel map:

     ros2 run go2_interfaces voxel_map_saver

   OPTIONAL Terminal 5 (Rosbag Recording):

   • If you want to record the data to a rosbag, use the following command to record the selected topics:

     ros2 bag record /tf /pointcloud /utlidar/robot_odom /camera/compressed

     Warning: Make sure your computer has enough space for recording, and do not record for too long to avoid filling up the disk. You can stop the recording by pressing Ctrl + C and you need to download it from the container.

   OPTIONAL Terminal 6 (save the map)

   • Important: You must wait for the scanning to finish before saving the map. AND PLEASE CHANGE THE PATH!!!!!!!!!!!!!!!!
   • Once the scanning is complete, you can save the voxel map using the following service call:

     ros2 service call /save_voxel_cloud go2_interfaces/srv/SaveVoxelCloud "{filename: '/YOUR FILE PATH/export.pcd'}"

⚠️⚠️⚠️⚠️⚠️ MAKE SURE TO RESTART THE ROBOT BEFORE RESTARTING THE CLIENT (HANDLE IT BEFORE RESTARTING AS IT WILL FALL DOWN)!!!!!!