Use this link to cite:
http://hdl.handle.net/2183/39717 Ferramenta para crear e supervisar misións en ROS para robots móbiles localizables con precisión
Loading...
Identifiers
Publication date
Authors
Montenegro Macía, Omar
Advisors
Other responsabilities
Universidade da Coruña. Facultade de Informática
Journal Title
Bibliographic citation
Type of academic work
Academic degree
Abstract
[Resumen]: En este proyecto se han analizado las herramientas existentes para definir y gestionar misiones de un robot móvil en ROS Melodic. De todas ellas, se ha escogido ROSMC por ser la más completa en cuanto a requisitos y la más flexible, escalable y versátil. Además, permite añadir la concurrencia de diferentes acciones y sincronizar la ejecución de misiones entre varios robots. Se han implementado nuevas acciones y se ha configurado ROSMC para gestionar un amplio abanico de misiones. Se han definido tres misiones tipo como referencia. La primera en interiores sirve para comprobar las funciones de navegación básicas del robot para planificar y ejecutas rutas con la presencia de obstáculos. La segunda misión pone a prueba la navegación en exteriores usando el GPS en ausencia de referencias del entorno. La tercera misión sincroniza una ruta con una acción (la captura continua de las imágenes de las cámaras). Se ha validado el sistema en entornos simulados y reales, tanto en entornos amigables en interiores como en entornos más desafiantes en exteriores. Como robot móvil se ha usado el Summit_XL equipado con varias cámaras y un lidar 3D. Los resultados de las pruebas demuestran la bondad de la herramienta diseñada. Aunque las pruebas reales en exteriores han supuesto un enorme desafío porque hemos necesitado reconfigurar el Summit_XL para poder operar con el GPS. Por último, comentar que para la navegación en exteriores se han estudiado diferentes tecnologías de mapas geolocalizados que sirvan al usuario como referencia para diseñar las rutas y misiones. Hemos añadido a ROSMC la integración con OpenStreetMap a través de rviz_satellite. Se ha elegido OpenStreetMap ya que permite guardar los mapas generados en caché para su uso offline, además de ser de uso libre y no requerir licencia. También permite añadir, configurar y adaptar diferentes capas a las necesidades del usuario final.
[Abstract]: In this project, existing tools for defining and managing missions for a mobile robot in ROS Melodic have been analyzed. Among all the options, ROSMC was chosen for being the most complete in terms of requirements and the most flexible, scalable, and versatile. Additionally, it allows for the concurrency of different actions and the synchronization of mission execution among multiple robots. New actions have been implemented, and ROSMC has been configured to manage a wide range of missions. Three reference missions have been defined as examples. The first one, for indoor environments, is used to verify the robot’s basic navigation functions for planning and executing routes in the presence of obstacles. The second mission tests outdoor navigation using GPS in the absence of environmental references. The third mission synchronizes a route with an action (continuous capture of images from the cameras). The system has been validated in both simulated and real environments, in friendly indoor settings as well as more challenging outdoor environments. The Summit_XL mobile robot, equipped with several cameras and a 3D lidar, was used. The test results demonstrate the effectiveness of the designed tool. However, real outdoor tests posed a significant challenge as we needed to reconfigure the Summit_XL to operate with GPS. Lastly, for outdoor navigation, different geolocated map technologies were studied to provide users with references for designing routes and missions. We integrated OpenStreetMap with ROSMC through rviz_satellite. OpenStreetMap was chosen because it allows generated maps to be cached for offline use, is free to use, and does not require a license. It also enables the addition, configuration, and adaptation of different layers to meet the needs of the end user.
[Abstract]: In this project, existing tools for defining and managing missions for a mobile robot in ROS Melodic have been analyzed. Among all the options, ROSMC was chosen for being the most complete in terms of requirements and the most flexible, scalable, and versatile. Additionally, it allows for the concurrency of different actions and the synchronization of mission execution among multiple robots. New actions have been implemented, and ROSMC has been configured to manage a wide range of missions. Three reference missions have been defined as examples. The first one, for indoor environments, is used to verify the robot’s basic navigation functions for planning and executing routes in the presence of obstacles. The second mission tests outdoor navigation using GPS in the absence of environmental references. The third mission synchronizes a route with an action (continuous capture of images from the cameras). The system has been validated in both simulated and real environments, in friendly indoor settings as well as more challenging outdoor environments. The Summit_XL mobile robot, equipped with several cameras and a 3D lidar, was used. The test results demonstrate the effectiveness of the designed tool. However, real outdoor tests posed a significant challenge as we needed to reconfigure the Summit_XL to operate with GPS. Lastly, for outdoor navigation, different geolocated map technologies were studied to provide users with references for designing routes and missions. We integrated OpenStreetMap with ROSMC through rviz_satellite. OpenStreetMap was chosen because it allows generated maps to be cached for offline use, is free to use, and does not require a license. It also enables the addition, configuration, and adaptation of different layers to meet the needs of the end user.
Description
Editor version
Rights
Atribución 3.0 España








