Metodología integrada para la implementación y visualización de datos LiDAR en vehículos autónomos mediante ROS y Jetson Nano

Abstract

This work presents a clear and replicable methodology for the utilization of LiDAR sensors in projects related to autonomous vehicles and Advanced Driver Assistance Systems (ADAS). It addresses the limitations of proprietary tools provided by sensor manufacturers by offering a flexible and customizable alternative through the Robot Operating System (ROS) and tools such as RVIZ. The methodology details the use of the SICK S3000 sensor in combination with Jetson Nano hardware, highlighting its capability to handle advanced libraries such as OpenCV, YOLO, and sick_scan. Key steps include configuration, data capture, storage in ROS Bag files, and conversion to more manageable formats like CSV, enabling offline analysis and experiment reproducibility. Practical examples of data visualization and results in vehicular contexts are presented, along with electrical configuration proposals to ensure the proper functioning of the sensor and complementary hardware. This work significantly contributes to the advancement of LiDAR sensor integration and application, providing accessible tools for students and