| KEYWORD |
Anomaly Detection on Italian Viaducts using Structural Health Monitoring Data
Tesi esterna in azienda
Parole chiave ACCELEROMETERS, ARTIFICIAL INTELLIGENCE, DEEP LEARNING, DEEP NEURAL NETWORKS, INCLINOMETERS, SOFTWARE, STRUCTURAL HEALTH MONITORING
Riferimenti ALESSIO BURRELLO, DANIELE JAHIER PAGLIARI
Riferimenti esterni Dr. Monica Longo (Sacertis S.r.l.)
Gruppi di ricerca DAUIN - GR-06 - ELECTRONIC DESIGN AUTOMATION - EDA, ELECTRONIC DESIGN AUTOMATION - EDA, GR-06 - ELECTRONIC DESIGN AUTOMATION - EDA
Tipo tesi EMBEDDED SOFTWARE DEVELOPMENT, EXPERIMENTAL, RESEARCH, SOFTWARE DEVELOPMENT
Descrizione During the thesis, the candidates will develop machine and deep learning models for anomaly detection on Italian bridges, utilizing inclinometer data from a Structural Health Monitoring (SHM) installations. The development of the thesis will be divided into three phases:
- Analysis of the models developed in the state of the art. In particular, we will examine both supervised and unsupervised algorithms to determine if a high-performing model can be achieved without training labels.
- Application of advanced automatic neural architecture search (NAS) developed within the research group to optimize the network’s performance.
The final outcome of the thesis will be a convolutional network, either supervised or unsupervised (e.g., autoencoder), to address a novel task using proprietary inclinometer data. Successfully addressing this task could improve privacy (eliminating the need for visual data) and lower installation costs, as it leverages pre-existing inclinometer data from SHM systems.
Conoscenze richieste Proficiency in Python is required. Familiarity with Deep Learning and the PyTorch library is a plus.
Note This thesis is offered in collaboration with Sacertis S.R.L., with the possibility for the candidate to conduct the work as part of an internship.
Scadenza validita proposta 04/11/2025
PROPONI LA TUA CANDIDATURA