Project Echo

Overview

Project Echo is a ROS 2-based robotics simulation project featuring a 3D-modeled robot integrated with Gazebo Classic and RViz for realistic simulation of perception and motion. The robot is equipped with a LiDAR scanner, enabling mapping with SLAM Toolbox and autonomous navigation using the Nav2 stack. This project showcases a complete simulation workflow, including teleoperation via keyboard control, SLAM-based mapping, and goal-directed navigation.

I developed this project to serve as a learning platform for:

• 3D robot modelling/urdf and xacro development
• Structuring effective launch files and ROS folder structure
• Integration of various ROS - Gazebo plugins for control methods and sensor perception
• Simulation of sensors for mapping and localization

Features

✅ 3D robot model visualized in RViz and Gazebo Classic

✅ LiDAR integration for environmental sensing

✅ SLAM Toolbox for real-time map building

✅ Nav2 for autonomous goal-based navigation

✅ Teleoperation via teleop_twist_keyboard

✅ Configurable launch system for simulation and RViz

Dependencies

Please ensure you have installed the necessary dependancies before getting started.

• Ubuntu 22.04
• ROS 2 Humble Hawksbill
• gazebo_ros
• xacro
• Slam ToolBox
• Nav2 Stack
• teleop_twist_keyboard
• xterm

Repository Structure

The following folders within the "echo_sim_v1" package are essential.

echo_sim_v1/
├── launch/           # Launch files (main, robot spawn, RViz)
├── urdf/             # Robot URDF and XACRO files
├── worlds/           # Gazebo world files
├── rviz/             # RViz configuration files
├── config/           # Navigation, SLAM, and sensor configs
├── meshes/           # 3D model meshes (STL)
└── README.md

Getting Started

1. Clone the Repo

git clone <your-repo-url>
cd echo_sim_v1

2. Build Workspace

colcon build
source install/setup.bash

3. Launch Simulation

Although there are several launch files in the launch directory (for initial testing purposes), the one that will be used is "gazebo.launch.py" which inherently calls "robot_spawn.launch.py"

The command below starts gazebo, spawns the robot in the world specified, opens Rviz with the saved config settings for laserscan messages and map saving, and launches teleop control:

ros2 launch echo_sim_v1 gazebo.launch.py   world:=/home/cameron/Project_Echo/Educational/amr_ws/src/echo_sim_v1/worlds/obstacles_1.world use_sim_time:=true

4. Launch SLAM

In a new terminal window, source the terminal and run the following command. This will launch the slam_toolbox and enable mapping of the world with the Lidar sensor. You can drive the robot around manually and begin mapping the virtual environment.

ros2 launch slam_toolbox online_async_launch.py params_file:=./src/echo_sim_v1/config/mapper_params_online_async.yaml use_sim_time:=true

5. Launch Nav2

In one last terminal window, source the terminal and run this command. This will launch Nav2 and allow you to set positinal goals in Rviz and let Nav2 plan a path for the robot autonomously.

ros2 launch nav2_bringup navigation_launch.py use_sim_time:=true

Map vs Virtual Environment

Notes:

• For teleop control, ensure your cursor is clicked on the xterm window before you use the keystrokes listed to control the robot.
• For mapping with SLAM, wait a few seconds after launching gazebo before launching the slam_toolbox. After you see the map fault go away on the Rviz side pane, you can start moving the robot around to map the environment.
• For autonomous control, select the "2D Goal Pose" option on the top of the Rviz menu. Then click on a point on the map you generated for the robot to begin its path planning.
• The simulation uses use_sim_time: true to ensure ROS time is synchronized with Gazebo.
• Ensure the robot frames map --> odom --> base_link are broadcasted properly or there may be TF extrapolation errors. Use TF tools to verify this.

Future Improvements:

• In the future I want to add a stereo camera as well as IMU sensor for improved perception of the virtual environment.
• I also want to improve the Nav2 functionality as there are still some small bugs I have encountered when testing.

References:

I referenced the youtube series: https://www.youtube.com/playlist?list=PLunhqkrRNRhYAffV8JDiFOatQXuU-NnxT by Articulated Robotics. This resource was extremely helpful.