Publication: Desarrollo de un sistema de procesamiento de imagen para un robot mini-humanoide basado en Model-in-the-Loop
Loading...
Identifiers
Publication date
2016-09
Defense date
2016-10-03
Authors
Tutors
Journal Title
Journal ISSN
Volume Title
Publisher
Impact
Export
Abstract
En este proyecto se ha desarrollado un sistema de procesamiento de imagen basado en modelos para un robot mini-humanoide con el objetivo de participar en el concurso nacional CEABOT. Se han aplicado los principios de visión por computador para adquirir, procesar y analizar una imagen del entorno del robot mini–humanoide RAIDER (Robot Antropomórfico para la Investigación y Desarrollo en Entornos Reales). Para ello se ha utilizado una webcam que ha permitido emular las imágenes que capturará el robot.
Las simulaciones que se han realizado en este proyecto, aplicando diferentes degradaciones en la imagen, intentan mostrar una aproximación a las distintas situaciones que se puede encontrar el robot en la realidad.
Se han utilizado las herramientas de MATLAB y Simulink, empleando la Computer Vision Toolbox que permitirá en un futuro modificar y ampliar el sistema, adaptándolo a las características específicas de cada cámara y robot.
En este documento se presentan las bases de la visión por computador, y posteriormente un análisis del procesamiento de imagen y de los algoritmos de obtención de distancias que han sido utilizados para poder realizar el desarrollo del sistema. Finalmente se muestran sus resultados.
An image processing system model has been developed for a humanoid robot, with the objective of participating in the national contest CEABOT. Computer vision principles have been applied in order to acquire, process and analyse the images of the environment around the humanoid robotic platform RAIDER (Anthropomorphic Robot for the Investigation and Development on Real Environments). A webcam has been used to emulate the capture of images that the robot would perform. Image degradations have been applied in the simulations to approximate the different environmental conditions that the robot would encounter as well. The software tools that have been used are Matlab and Simulink, including the Computer Vision Toolbox. Thanks to these tools, it will be possible to modify or adapt the developed image processing system to the characteristics of each camera and robot. This document introduces the bases of computer vision, also presenting an analysis of the different ways to process an image and the algorithms used to calculate distances. The results of this research have been used to develop the final system. As a final point, the results are presented.
An image processing system model has been developed for a humanoid robot, with the objective of participating in the national contest CEABOT. Computer vision principles have been applied in order to acquire, process and analyse the images of the environment around the humanoid robotic platform RAIDER (Anthropomorphic Robot for the Investigation and Development on Real Environments). A webcam has been used to emulate the capture of images that the robot would perform. Image degradations have been applied in the simulations to approximate the different environmental conditions that the robot would encounter as well. The software tools that have been used are Matlab and Simulink, including the Computer Vision Toolbox. Thanks to these tools, it will be possible to modify or adapt the developed image processing system to the characteristics of each camera and robot. This document introduces the bases of computer vision, also presenting an analysis of the different ways to process an image and the algorithms used to calculate distances. The results of this research have been used to develop the final system. As a final point, the results are presented.
Description
Keywords
Proceso de imágenes, Visión artificial, Robótica, Robot mini humanoide