Adaptive Clustering for 2D LiDAR in ROS

Overview

This ROS package implements adaptive clustering for 2D LiDAR data. The adaptive clustering algorithm groups nearby points in the LiDAR scan based on adaptive thresholds, providing a more accurate representation of objects in the environment.

Demo

Topics

• ROS
• LiDAR
• Adaptive Clustering
• Robotics
• Sensor Data
• Visualization
• Python
• Catkin
• Marker Array

Tags

ROS, LiDAR, Adaptive Clustering, Robotics, Sensor Data, Visualization, Python, Catkin, Marker Array

Prerequisites

• ROS (Robot Operating System)
• Catkin workspace

Installation

1. Clone the repository to your catkin workspace:

   git clone https://github.com/your-username/adaptive_clustering_2D.git

2. Build the catkin workspace:

   cd path/to/your/catkin_workspace
   catkin_make
   source devel/setup.bash

Usage

Launching the Adaptive Clustering Node

To launch the adaptive clustering node with default settings (subscribing to the /scan topic):

roslaunch adaptive_clustering_2D adaptive_clustering.launch

To launch the adaptive clustering node with a custom scan topic (e.g., /my_custom_scan):

roslaunch adaptive_clustering_2D adaptive_clustering.launch scan_topic:=/my_custom_scan

Parameters

• scan_topic (default: /scan): The LiDAR scan topic to subscribe to.
• cluster_size_min (default: 3): Minimum size of a cluster to be considered.
• tolerance_factor (default: 0.1): Factor used to calculate adaptive threshold.

Adjust these parameters in the launch file as needed.

Citations

@article{gomez2023efficient,
title={Efficient Detection and Tracking of Human Using 3D LiDAR Sensor},
author={G{'o}mez, J. and Aycard, O. and Baber, J.},
journal={Sensors},
volume={23},
number={10},
pages={4720},
year={2023},
}