ROS2 Camera-LiDAR Fusion

A ROS2 package for calculating intrinsic and extrinsic calibration between camera and LiDAR sensors. This repository provides an intuitive workflow to fuse data from these sensors, enabling precise projection of LiDAR points into the camera frame and offering an efficient approach to sensor fusion.

Visual Overview

| Static Sensors | Moving Sensors | |———————|——————–| | | |

Support 💖

If you find this project helpful and want to support its ongoing development, you can buy me a coffee! Every contribution helps me dedicate more time to improving and maintaining open-source software.

Table of Contents

1. Get Started
   • Prerequisites
   • Installation
2. Usage
   • Node Overview
   • Workflow
   • Running Nodes
3. License

Get Started

Prerequisites

To run this package, ensure the following dependencies are installed:

• Git: For version control and repository management.
• Docker: To streamline the environment setup and execution.
• NVIDIA Container Toolkit (if using an NVIDIA GPU): For hardware acceleration.

Installation

Clone the Repository

Start by cloning the repository:

git clone git@github.com:CDonosoK/ros2_camera_lidar_fusion.git

Build Using Docker

This repository includes a pre-configured Docker setup for easy deployment. To build the Docker image:

1. Navigate to the docker directory:

   cd ros2_camera_lidar_fusion/docker
   

1. Run the build script:

   sh build.sh
   

This will create a Docker image named ros2_camera_lidar_fusion.

Run the Docker Container

Once built, launch the container using:

sh run.sh

Usage

Node Overview

This package includes the following ROS2 nodes for camera and LiDAR calibration:

Workflow

Follow these steps to perform calibration and data fusion:

1. Intrinsic Calibration
   Run get_intrinsic_camera_calibration.py to generate the intrinsic calibration file for the camera.

2. Data Recording
   Use save_sensor_data.py to capture and save synchronized data from the camera and LiDAR.

3. Point Correspondence
   Execute extract_points.py to manually select corresponding points between camera and LiDAR.

4. Extrinsic Calibration
   Run get_extrinsic_camera_calibration.py to compute the transformation matrix between camera and LiDAR.

5. LiDAR Projection
   Use lidar_camera_projection.py to project LiDAR points into the camera frame for visualization and analysis.

Running Nodes

To execute a specific node, use the following command:

ros2 run ros2_camera_lidar_fusion <node_name>

For example:

ros2 run ros2_camera_lidar_fusion lidar_camera_projection.py

Maintainer

This package is maintained by:

Clemente Donoso
Email: clemente.donosok@gmail.com GitHub: CDonosoK

License

This project is licensed under the MIT. See the LICENSE file for details.

Contributions and feedback are welcome! If you encounter any issues or have suggestions for improvements, feel free to open an issue or submit a pull request.