This is a simple bot that can autonomously navigate unknown environments. It has a lidar, camera and wheel encoders for sensing and a differential drive for movement.

Necessary dependencies:

• ROS2 Humble (Developed in this version)
• Gazebo classic
• Navigation2
• SLAM Toolbox
• xacro
• gazebo_ros_pkgs
• twist_mux
• ros2_control

To install the package, run the following commands:

mkdir -p ~/ws_my_bot/src && cd ~/ws_my_bot/src

git clone https://github.com/PrakritTyagi/my_bot.git 

cd ~/ws_my_bot

colcon build

source install/setup.bash

To launch the gazebo simulation, run the following command:

ros2 launch my_bot display.launch.py 

To launch the slam tool, run the following command:

ros2 launch my_bot online_async_launch.py use_sim_time:=true

To launch the navigation tool, run the following command:

ros2 launch nav2_bringup navigation_launch.py use_sim_time:=true

The simulation world is duckytown.world. It is a simple world in urban environment with a few obstacles.

The docker file is provided in the my_bot/dockerfile directory. You can copy it to the root of the workspace and build the docker image. I used this with vscode remote container extension.

• If while running docker the gazebo takes time to load, change the world file to a simple one. In my experience, you would have to wait a while for the gazebo to load the duckytown.world file.