• 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
 
 
Doctoral Thesis
DOI
https://doi.org/10.11606/T.18.2021.tde-27052022-103103
Document
Author
Full name
João Paulo Brognoni Casati
Institute/School/College
Knowledge Area
Date of Defense
Published
São Carlos, 2021
Supervisor
Committee
Altafim, Ruy Alberto Corrêa (President)
Amorim, Mardson Freitas de
Fernandes, Ricardo Augusto Souza
Flauzino, Rogério Andrade
Spatti, Danilo Hernane
Title in Portuguese
Análise e simulação de sinais de acelerômetro gerados por vibração de impacto de veículos em lombada utilizando inteligência artificial
Keywords in Portuguese
Deep learning
Acelerômetro
Aprendizagem de máquina
Cidades inteligentes
Geração de dados
Inteligência artificial
Processamento de sinais
Redes neurais artificiais
Abstract in Portuguese
Analisar veículos e seu comportamento pode ser útil para o monitoramento de diversos problemas de tráfego, como danos à pavimentação e engarrafamentos, assim como a classificação de tipo de veículo. O objetivo deste trabalho é realizar uma análise aprofundada do comportamento da vibração causada pelo impacto de veículos em um obstáculo utilizando inteligência artificial. Para a análise destes dados são utilizadas técnicas de deep learning que possibilitam a utilização dos 3 eixos de dados do acelerômetro de forma crua, além da simulação de amostras artificiais que permitem melhorar os resultados e resolver problemas de oversampling. Conclui-se com este trabalho que os resultados alcançados são promissores e permitem avanços em aplicações que possam considerar tráfego de veículos, monitoramento e soluções em cidades inteligentes.
Title in English
Analysis and simulation of accelerometer signals generated by vehicle impact vibration in speed bump using artificial intelligence
Keywords in English
Accelerometer
Artificial intelligence
Artificial neural networks
Data augmentation
Deep learning
Machine learning
Signal processing
Smart cities
Abstract in English
Analyzing vehicles and their behavior can be useful for monitoring various traffic problems, such as damage to pavement and traffic jams, as well as vehicle type classification. The objective of this work is to carry out an in-depth analysis of the vibration behavior caused by the impact of vehicles on an obstacle using artificial intelligence. For the analysis of this data, deep learning techniques are used, which allow the raw use of the 3 axes of the accelerometer data, in addition to the simulation of synthetic samples that improve the results and solve oversampling problems. This work concludes that the results achieved are promising and allow advances in applications conidering vehicle traffic, monitoring and solutions in smart cities.
 
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
2022-05-27
 
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.