ROS Simple Sample Packages

This repo has some ROS components to help you initialize your own ROS package, depending on what you aim to do.

User-Created Packages

• beginner_tutorials:   The completed beginner section of official ROS tutorials. This is the only package in this repo written in C++.
• fundamentals:   Fundamentals for writing a publisher, subscriber, service, client, action service, and action client.
• labyrinth:   Launch a turtlebot3 inside a labyrinth world modeled from scratch using Gazebo.
• simple_applications:   Move turtlebot3 in an empty world to achieve certain goals, such as going in a straight line, drawing a circle, or patrolling back and forth. Also includes files to parse camera and LIDAR data from onboard sensors.
• gui_control:   Use a GUI that is designed from scratch to control turtlebot3 in different worlds. Consider going over the PyQt5 fundamentals before visiting this package.
• opencv_img_processing:   Includes OpenCV image processing fundamentals, a demo webcam application to extract blue-colored objects, and a simple path tracking example for turtlebot3.
• slam_and_navigation:   Apply localization, mapping, and autonomous navigation for turtlebot3 using existing packages in ROS. Waypoint placement for navigation is explained using both methods: using RViz or ROS actions.
• robot_modelling:   Model a robot from scratch that is dissimilar to turtlebot3, but also accommodates two regular wheels, two caster wheels, an onboard camera and LIDAR. Observe the robot in 3D through Gazebo.

Notes

• If you're using an IDE, open your workspace root directory as the catkin_ws folder in this repo, as named commonly in the ROS community.
• If you get errors during catkin_make, especially after moving directories, delete the compiled build, devel and install folders in catkin_ws and re-create them using catkin_make and catkin_make install.
• Remember to source catkin_ws/devel/setup.sh for every new terminal. If frequently used, consider adding this command to ~/.bashrc.
• Some developers prefer to place C++ code into a src folder, and Python code into a scripts folder instead. Throughout this repo, only a src folder is used, regardless of the language.
• At the very top of each Python script, you should have a preamble similar to #!/usr/bin/env python3 that tells the executor where python3 is located.
• You will need to one-time chmod +x each Python script after creating it, making it executable. Once done, you should see the filename show as green instead of white when you ls.
• The packages related to turtlebot3 inside catkin_ws/src were cloned directly from their Git repos using the following commands (remember to catkin_make your workspace after cloning).

cd catkin_ws/src
git clone https://github.com/ROBOTIS-GIT/turtlebot3_msgs.git
git clone https://github.com/ROBOTIS-GIT/turtlebot3.git -b noetic-devel
git clone https://github.com/ROBOTIS-GIT/turtlebot3_simulations.git -b noetic-devel

• For anything related to turtlebot3, you will need to export TURTLEBOT3_MODEL=<bot_type> in each terminal. Options for <bot_type> are burger, waffle and waffle_pi. Do not add any space before or after the = symbol. If frequently used, consider adding this command to ~/.bashrc. Applications in this repo assumes waffle_pi is exported.
• For SLAM and navigation applications, you need to install the following packages.

sudo apt-get install ros-noetic-gmapping
sudo apt-get install ros-noetic-navigation

• To save the tree of links and joints into a PDF file after robot modelling, you need to install the following package.

sudo apt-get install ros-noetic-tf2-tools

Cheatsheet

• For subscribers, spin() just lets all the callbacks get called in an infinite-loop, and is not used for publishers since there is no callback function. For publishers, just enter a while loop and publish.
• The command rospy.Rate(foo) is only to modify the publishing rate, and therefore is only used for publishers. You do not need to use it for subscribers.
• Why do we use rospy.spin() for subscribers? From the ROS Wiki: The simplest (and most common) version of single-threaded spinning is ros::spin(). In this application all user callbacks will be called from within the ros::spin() call. ros::spin() will not return until the node has been shutdown, either through a call to ros::shutdown() or a Ctrl-C.
• We use rate.sleep() to pause the script in order to hit the hz mark of rate = rospy.spin(foo). The time past since the last entry to rate.sleep() is computed under-the-hood, and the corresponding sleep-time is executed.

References

This repo borrows a significant amount of material from this Udemy course (Turkish), by Muhammed Oğuz Taş. Consider supporting him by purchasing his course on Udemy if you have found this repo to be helpful.