Concepts and Packages

ROS 1

Core Concepts of ROS 1

Playlist encapsualting important concepts

Tutorials: https://wiki.ros.org/ROS/Tutorials

ROS 1 is structured around several key concepts that enable modular, distributed, and scalable robot software development. The main ideas are organized at different levels25:

1. Nodes

• Definition: Nodes are independent processes that perform computation. Each node is typically responsible for a single task (e.g., reading sensor data, controlling motors, localization).

• Implementation: Nodes are written using ROS client libraries such as roscpp (C++) or rospy (Python).

• Communication: Nodes communicate with each other using topics, services, or the parameter server, and are managed at runtime by the ROS Master235.

2. Master

• Role: The ROS Master provides name registration and lookup for nodes, topics, and services, enabling nodes to discover and connect with each other.

• Operation: Nodes register themselves and their communication interfaces with the Master, which acts like a DNS server for the ROS network. After discovery, nodes communicate directly25.

3. Parameter Server

• Purpose: Stores global parameters (key-value pairs) accessible at runtime by all nodes. Used for configuration and tuning of nodes without code changes25.

4. Topics

• Mechanism: Topics are named message buses for asynchronous, many-to-many communication. Nodes publish messages to topics, and other nodes subscribe to receive those messages.

• Strong Typing: Each topic enforces a specific message type for data consistency.

• Transport: Default is TCPROS (reliable TCP/IP), with UDPROS (UDP-based, low-latency) for specific use cases235.

5. Messages

• Definition: Strictly defined data structures used to communicate over topics. Defined in .msg files and can include primitive types or nested messages35.

6. Services

• Mechanism: Synchronous, request/response interactions for tasks that require confirmation or a result (e.g., requesting a sensor reading).

• Definition: Defined in .srv files with request and response fields5.

7. Bags

• Purpose: Files (with .bag extension) for recording and replaying topic data. Essential for debugging, simulation, and development5.

8. Filesystem Level

• Packages: The basic unit of ROS software organization, containing nodes, libraries, configuration files, and more.

• Workspaces: Directories for building and managing multiple packages together5.

Popular and Essential ROS 1 Packages

While the importance of a package depends on your application, the following are widely used and considered essential in most ROS 1 projects467:

Package Name	Purpose/Description
tf / tf2	Coordinate frame transforms (essential for robot kinematics)
rviz	3D visualization tool for sensor data, robot models, and planning
ros_control	Framework for controller management and hardware abstraction
moveit	Motion planning and manipulation for robotic arms
navigation	Autonomous navigation stack for mobile robots
gazebo_ros_pkgs	Integration between ROS and Gazebo simulator
urdf	Unified Robot Description Format for robot models
robot_localization	State estimation from multiple sensors
rosbag	Data recording and playback
sensor_msgs, geometry_msgs, std_msgs	Standard message types for sensors, geometry, etc.
image_transport, cv_bridge	Tools for handling images and computer vision
robot_state_publisher	Publishes robot joint states to tf
actionlib	Tools for asynchronous task execution
ros_perception	Perception stack for vision and point cloud processing
rosparam, rosnode, rostopic, rosservice	Core tools for managing ROS system
ROS Industrial	Packages for industrial robot support
rqt	GUI plugins for introspection, plotting, and debugging

Other Notable Packages (from usage statistics)6:

• cv_bridge, image_transport (vision)

• robot_state_publisher, joint_state_publisher (robot state)

• dynamic_reconfigure (runtime parameter tuning)

• catkin (build system)

• roslaunch, rosbag, rosmsg, rostopic, rosservice, rosnode (core utilities)

• urdf_tutorial, turtlesim (learning and demos)

References and Further Reading

• ROS Concepts (wiki.ros.org)2
• The Building Blocks of ROS 1 (Foxglove)3
• Effective Robotics Programming with ROS (Packt)5
• Most Important ROS Packages (Reddit)4
• Top ROS Packages (GitHub Gist)6
• Trossen Robotics: ROS 1 Open Source Packages

ROS 2

Playlist encapsualting important concepts

ROS 2 Concepts and Popular Packages

ROS 2 Tutorials: https://docs.ros.org/en/humble/Tutorials.html

Core Concepts of ROS 2

ROS 2 is a modern, distributed robotics middleware designed for flexibility, scalability, and real-time performance. It builds on the lessons of ROS 1 and introduces several key architectural and functional improvements.

1. Nodes

• Definition: Nodes are independent, modular processes that perform computation. Each node can handle a specific function, such as sensor reading, data processing, or actuator control.

• Types:

  • Standard Node: Basic node with minimal state management.

  • Lifecycle Node: Supports managed states (e.g., unconfigured, inactive, active, finalized), enabling robust startup, shutdown, and error handling47.
• Features: Nodes are more modular in ROS 2, can be composed into a single process, and support advanced lifecycle management for reliability47.

2. Topics

• Mechanism: Topics are named channels for asynchronous, many-to-many publish/subscribe communication. Nodes publish messages to topics, and other nodes subscribe to receive them246.
• DDS Integration: ROS 2 uses the Data Distribution Service (DDS) middleware, enabling automatic discovery, decentralized communication, and advanced Quality of Service (QoS) controls67.
• QoS Policies: Fine-grained control over reliability, durability, deadline, and resource usage, crucial for real-time and safety-critical applications47.

3. Services

• Definition: Services provide synchronous, request/response communication. A client sends a request to a service server and waits for a response, suitable for tasks that require confirmation or a result46.
• Implementation: Defined using .srv files specifying request and response types.

4. Actions

• Purpose: Actions allow for asynchronous, long-running tasks with feedback and the ability to cancel (e.g., navigation goals, manipulation tasks).

• Structure: Built on top of topics and services, actions provide goal, feedback, and result channels46.

5. Parameters

• Role: Store configuration and runtime values for nodes. Parameters can be dynamically updated, allowing for flexible system tuning46.

6. Quality of Service (QoS)

• Importance: QoS policies (enabled by DDS) let developers specify message delivery guarantees, reliability, and resource constraints for each topic or service, enhancing real-time and distributed performance47.

7. Launch System

• Advancement: ROS 2 uses Python-based launch files, allowing for programmable, flexible system startup and orchestration46.

8. Security

• Features: Built-in support for authentication, encryption, and access control, making ROS 2 suitable for commercial and safety-critical deployments47.

9. Tools and Ecosystem

• Developer Tools: ROS 2 offers a suite of tools for introspection, debugging, visualization (e.g., RViz2), plotting, logging, and playback, accelerating development and troubleshooting57.
• Community: A vibrant, global community contributes to a rich repository of open-source packages, ensuring continuous improvement and innovation7.

Popular and Essential ROS 2 Packages

The following packages are widely used and form the backbone of most ROS 2 applications. Your choice may vary depending on your robot and application, but these are considered foundational:

Package Name	Purpose/Description
rclcpp / rclpy	Core C++ and Python client libraries for node development
tf2	Coordinate frame transformations for robot kinematics and sensor fusion
RViz2	3D visualization tool for sensor data, robot models, and planning
ros2_control	Modular, extensible robot control framework
Nav2	Navigation stack for autonomous mobile robots
MoveIt 2	Advanced motion planning and manipulation for robotic arms
Gazebo / Ignition	High-fidelity robot simulation environments
sensor_msgs, geometry_msgs, std_msgs	Standard message types for sensors, geometry, and basic data
image_tools, camera_info_manager	Vision and sensor integration tools
rosbag2	Data recording and playback for ROS 2 messages
robot_state_publisher	Publishes robot joint states to tf2
action_msgs, lifecycle_msgs	Standard messages for actions and node lifecycle management
rqt, RViz2 plugins	GUI tools for introspection, plotting, debugging, and visualization
slam_toolbox	SLAM (Simultaneous Localization and Mapping) for mapping and localization
navigation2 (Nav2)	Advanced navigation and path planning for mobile robots
Perception packages	Includes packages for vision, point cloud processing, and object detection

Other Notable Packages:

• micro_ros (ROS 2 for microcontrollers)

• ament_cmake (build system)

• colcon (build tool)

• ros2launch (launch management)

• ros2param (parameter management)

• ros2topic, ros2service, ros2bag (core CLI tools)

References and Further Reading

• ROS 2 Concepts (docs.ros.org, Foxy)2
• ROS 2 Concepts (docs.ros.org, Humble)6
• Introduction to ROS 2 (LearnOpenCV)4
• ROS 2 Tutorials (Husarion)5
• Introduction to ROS 2 (MathWorks)7
• Most Important ROS Packages (Reddit)