ROS2 Action Client Server

A professional ROS2 Action communication project using a custom CountUntil.action interface with Action Client and Action Server architecture.

This project demonstrates:

• ROS2 Actions
• Custom Action Interfaces
• Action Server
• Action Client
• Goal Handling
• Result Processing
• Asynchronous Communication
• Real-time Execution Workflow
• Industry-style ROS2 Package Structure

---

Project Overview

This project implements a complete ROS2 Action workflow.

The Action Client sends a goal to the Action Server. The Action Server receives the goal and starts counting from 0 up to the target number with a delay between each count.

After execution is complete, the server returns the final result back to the client.

This project is useful for understanding:

• Long-running tasks in ROS2
• Why Actions are better than Services for long tasks
• Client-server communication in robotics
• Asynchronous execution
• Goal/result handling

---

ROS2 Action Workflow

Action Client  ----------------->  Action Server
      |                                |
      |-------- Send Goal ------------>|
      |                                |
      |<------- Goal Accepted ---------|
      |                                |
      |<-------- Feedback -------------|
      |                                |
      |<--------- Result --------------|

---

Custom Action Interface

File:

action/CountUntil.action

Action definition:

# Goal
int64 target_number
float64 delay
---
# Result
int64 reached_number
---
# Feedback
int64 current_number

---

Workspace Structure

action_ws/
│
├── src/
│   ├── my_robot_interface/
│   │   ├── action/
│   │   │   └── CountUntil.action
│   │   ├── CMakeLists.txt
│   │   └── package.xml
│   │
│   └── my_py_pkg/
│       ├── count_until_server.py
│       ├── count_until_client.py
│       ├── package.xml
│       └── setup.py
│
└── install/

---

Features

• Custom ROS2 Action Interface
• Action Server Implementation
• Action Client Implementation
• Goal Acceptance
• Goal Execution
• Result Handling
• Feedback Communication
• Asynchronous Execution
• Professional ROS2 Architecture

---

Installation

Clone Repository

git clone https://github.com/omkar-robotics/ros2-action-client-server.git

Go to Workspace

cd ros2-action-client-server/action_ws

Build Workspace

colcon build

Source Workspace

source install/setup.bash

---

Run the Action Server

Open Terminal 1:

ros2 run my_py_pkg count_until_server

---

Run the Action Client

Open Terminal 2:

ros2 run my_py_pkg count_until_client

---

Example Output

Server Output

[INFO] Action Server Started
[INFO] Received goal request
[INFO] Executing goal...
[INFO] Counting: 1
[INFO] Counting: 2
[INFO] Counting: 3
[INFO] Counting: 4
[INFO] Counting: 5
[INFO] Goal completed

Client Output

[INFO] Goal accepted
[INFO] Current Number: 1
[INFO] Current Number: 2
[INFO] Current Number: 3
[INFO] Current Number: 4
[INFO] Current Number: 5
[INFO] Result: 5

---

Important ROS2 Concepts Used

Topics

Used for continuous data streaming.

Example:

• Sensor data
• Camera feed
• Robot position

---

Services

Used for quick request-response communication.

Example:

• Turn LED ON/OFF
• Reset counter

---

Actions

Used for long-running tasks.

Actions support:

• Goal
• Result
• Feedback
• Cancel Request

Example:

• Robot Navigation
• Autonomous Movement
• Pick and Place Tasks

---

Why Actions are Important

ROS2 Services are not suitable for long-running tasks because:

No feedback mechanism No cancellation support Client waits blindly Difficult to monitor progress

ROS2 Actions solve these problems using:

Real-time feedback Goal management Asynchronous communication Result handling Goal cancellation support

This makes Actions ideal for robotics applications.

---

Technologies Used

• ROS2 Humble
• Python
• rclpy
• colcon
• Ubuntu Linux

---

Future Improvements

• Goal Cancellation Mechanism
• Multi-Threaded Executor
• Reentrant Callback Groups
• Multiple Goal Handling
• Fibonacci Action Example
• Robot Navigation Actions
• Gazebo Integration
• RViz Visualization
• Real Robot Hardware Integration

---

Learning Outcome

Through this project, I learned:

• ROS2 Action Architecture
• Custom ROS2 Interfaces
• Client-Server Communication
• Asynchronous Programming in ROS2
• Goal/Result/Feedback Workflow
• ROS2 Package Structure
• Professional Robotics Development Workflow

---

Author

Omkar Honrao

(Robotics research intern)

GitHub:

https://github.com/omkar-robotics