Banca de QUALIFICAÇÃO: MATEUS FILIPE SILVA

Uma banca de QUALIFICAÇÃO de MESTRADO foi cadastrada pelo programa.
DISCENTE : MATEUS FILIPE SILVA
DATA : 20/11/2025
HORA: 19:00
LOCAL: Sala Virtual:meet.google.com/cvg-eswh-nhj
TÍTULO:

 An Artificial Intelligence-Based Approach for Localization and Topological Navigation of Smart Vehicles.

PALAVRAS-CHAVES:

Topological localization, Computer vision, Machine learning, Intelligent vehicles, Autonomous navigation.

PÁGINAS: 47
GRANDE ÁREA: Engenharias
ÁREA: Engenharia Elétrica
RESUMO:

This work presents an artificial intelligence-based approach for the localization and topological navigation of intelligent vehicles, focusing on urban scenarios. The methodology employs computer vision and machine learning to replace manual descriptors with deep learning models, aiming for improved accuracy and computational efficiency compared to the SURF + KNN Match + RANSAC method proposed by Neto et al. (2024). The system consists of two models: the first identifies whether the vehicle is on a straight segment or at an intersection, and the second classifies the corresponding topological node on the map. The dataset was collected in a real urban environment and contains 3,132 georeferenced images. Preliminary results show accuracy above 90% in their respective tasks, suggesting performance exceeding the reference work and indicating the feasibility of real-time application.

MEMBROS DA BANCA:
Presidente - 2362923 - GIOVANI BERNARDES VITOR
Interno - ***.230.456-** - LUIZ CARLOS BAMBIRRA TORRES - UFOP
Interno - 1183761 - SANDRO CARVALHO IZIDORO