• JoomlaWorks Simple Image Rotator
  • JoomlaWorks Simple Image Rotator
  • JoomlaWorks Simple Image Rotator
  • JoomlaWorks Simple Image Rotator
  • JoomlaWorks Simple Image Rotator
  • JoomlaWorks Simple Image Rotator
  • JoomlaWorks Simple Image Rotator
  • JoomlaWorks Simple Image Rotator
  • JoomlaWorks Simple Image Rotator
  • JoomlaWorks Simple Image Rotator
 
  Bookmark and Share
 
 
Master's Dissertation
DOI
https://doi.org/10.11606/D.55.2016.tde-18082016-155035
Document
Author
Full name
Marcos Gonçalves Quiles
Institute/School/College
Knowledge Area
Date of Defense
Published
São Carlos, 2004
Supervisor
Committee
Romero, Roseli Aparecida Francelin (President)
Simões, Eduardo do Valle
Terra, Marco Henrique
Title in Portuguese
Sistema de visão baseado em redes neurais artificiais para o controle de robôs móveis
Keywords in Portuguese
Não disponível
Abstract in Portuguese
Sistemas de visão computacional (SVCs) são interessantes ferramentas para a navegação de robôs. Este trabalho propõe um SVC baseado em redes neurais multi-camadas para o controle de um robô móvel. Dois módulos principais compõe o SVC: módulo de visão e módulo de controle de navegação. O primeiro é dividido em três partes:pré-processamento, segmentação e reconhecimento das imagens. este módulo processa as imagens (obtidas por uma câmera) que podem conter diversos com cores diferentes, retornando a posição ou a forma e a posição de um dos objetos, o qual corresponde a cor especificada. O processamento das imagens é realizada por redes neurais multi-camadas. O módulo de controle é responsável por validar os resultados do módulo de visão e conduzir o robô utilizando os dados provenientes do módulo de visão. O objetivo do sistema proposto é capacitar um robô a realizar as tarefas: seguir um objeto de cor determinada, seguir um objeto de cor e forma determinada ou ainda navegar pelo ambiente seguindo um objeto de cor determinada evitando obstáculos. Vários experimentos são apresentados, em um ambiente real, utilizando o robô. Pioneer I, para mostrar as vantagens e desvantagens do sistema proposto.
Title in English
Vision system based on artificial neural networks for controlling mobile robots
Keywords in English
Not available
Abstract in English
Computer Systems Vision (CVSs) are important tools navigation. This work proposes a CVS based on Multi-Layer Neural Networks for a mobile robot control. Two main modules compose the CVS: vision module and navigation control module. The first is divided into three parts: preprocessing, segmentation and image recognition. This module processes the images (obtained by a camera) that can contain several objects with different colors, returning the position or the shape and position of one of these objects, which corresponds to a specified color. The image processing is carried out by multi-layer neural networks. The control module valides the results from the vision module and leads the robot using data from the vision module. the aim of the proposed system is to enable a robot to perform the tasks: to follow an object with specified color, to follow an object with a specified color and shape or even to navigate in an environment following and object with a specified color avoiding obstacles. Some experiments are presented in a real environment with the robot Pionner I in order to show the advantages and disvantages of proposed system.
 
WARNING - Viewing this document is conditioned on your acceptance of the following terms of use:
This document is only for private use for research and teaching activities. Reproduction for commercial use is forbidden. This rights cover the whole data about this document as well as its contents. Any uses or copies of this document in whole or in part must include the author's name.
Publishing Date
2016-08-19
 
WARNING: Learn what derived works are clicking here.
All rights of the thesis/dissertation are from the authors
CeTI-SC/STI
Digital Library of Theses and Dissertations of USP. Copyright © 2001-2024. All rights reserved.