Banca de QUALIFICAÇÃO: MORSINALDO DE AZEVEDO MEDEIROS

Uma banca de QUALIFICAÇÃO de MESTRADO foi cadastrada pelo programa.
STUDENT : MORSINALDO DE AZEVEDO MEDEIROS
DATE: 17/02/2025
TIME: 14:00
LOCAL: Remoto
TITLE:

MMCloud: an online unsupervised clustering algorithm for driver behavior analysis

KEY WORDS:

Vehicle sensors, Internet of Things (IoT), TinyML, real-time processing, Edge Computing, OBD-II Edge,

PAGES: 81
BIG AREA: Engenharias
AREA: Engenharia Elétrica
SUBÁREA: Eletrônica Industrial, Sistemas e Controles Eletrônicos
SPECIALTY: Automação Eletrônica de Processos Elétricos e Industriais
SUMMARY:

The significant increase in vehicle sensors, combined with the convergence of Internet of Things (IoT) technologies, has generated a large volume of data, enabling real-time analysis at the network edge through OBD-II Edge devices. This study proposes developing, implementing, and validating a real-time vehicle data processing solution to classify driver behavior. The approach is structured into five layers: vehicle data collection, soft-sensor calculation, execution of the MMCloud algorithm (an incremental online clustering method), cluster labeling, and generation of evaluation metrics. The methodology was integrated into low-energy and low-computational-cost hardware, utilizing TinyML techniques. A case study was conducted in Natal-RN, Brazil, with two participants using the Freematics One+ device. Preliminary results showed promise in classifying driver behavior, capturing significant nuances throughout the journey. It is concluded that the study has the potential to contribute to the classification of driver profiles, promoting safety and efficiency in traffic.

COMMITTEE MEMBERS:
Presidente - 2885532 - IVANOVITCH MEDEIROS DANTAS DA SILVA
Interno - 1153006 - LUIZ AFFONSO HENDERSON GUEDES DE OLIVEIRA
Externo à Instituição - IGNACIO SANCHEZ GENDRIZ - UFRN
Externa à Instituição - MARIANNE BATISTA DINIZ DA SILVA - UFAL