Teacher: Giovanna Venuti
Tutor: Andrea Gatti (Geomatics Research and Development srl – GReD)
Research area: GNSS, Geospatial data analysis, Artificial intelligence
Description: The project focuses on the development and evaluation of deep learning models, including transformer architectures, for anomaly detection in GNSS-based displacement time series collected by structural and environmental monitoring systems. The work will investigate methods to identify abnormal behaviour in 3D displacement signals and explore predictive approaches to detect changes in deformation trends relevant for early warning applications. The analysis may also include station telemetry (power, communications, estimation quality) to detect system malfunctions or performance degradation.
Keywords: GNSS, anomaly detection, deep learning
Technologies: Implementation will be carried out in Python using common deep learning libraries (e.g., PyTorch or TensorFlow).