and the automated search and removal of anomalous solutions. Recently, interest has been devoted to the problem of the automated detection of discontinuities in GNSS time series by developing a new algorithm called STARS-3D.
Still related to the use of GNSS technology, in the field of structural monitoring, the low-cost receivers recently introduced on the market were studied and tested, and an optimised approach was developed for the case of monitoring tall structures in an urban environment. In the same context (monitoring of the Garisenda tower in Bologna), an approach was developed to increase the sensitivity of the monitoring system to anomalous movements in real time.
The research activity was mainly focused on topics in the areas of Operational Geodesy, satellite surveying, and structure and land monitoring, with particular reference to GNSS technology, but not only. A first area of research relates to the calculation strategies of large networks of GNSS permanent stations, which constitute the fundamental geodetic infrastructure for the definition and monitoring of modern reference frames, both on a global and regional scale. Particular attention was initially devoted to the pre-analysis phase of the data archives and the calculation of the National Dynamic Network, the current official Italian reference frame.
In-depth research was conducted on the technique for calculating GNSS data observables called PPP (Precise Point Positioning), implemented in the GIPSY OASIS II software (JPL - NASA) and which has become a standard in the field of Geodesy and Solid Earth Geophysics in recent years. Studies have been conducted on methods for framing PPP solutions on international ITRS-ETRS reference systems, including through the calculation of ad hoc transformation parameters. The real-time PPP positioning technique was researched in a test performed with commercial Atlas technology.
Machine Learning algorithms applied to GNSS time series were researched in order to reduce measurement noise and improve the interpretability of the investigated phenomena. Machine Learning techniques were also applied to the problem of super-resolution of thermal images with ‘single frame’ approaches, while an ad hoc algorithm was developed for super-resolution from multiple images.
The problem of monitoring the Emilia-Romagna coastline (conducted in collaboration with the Arpae sea-coast unit), also through the use of photogrammetric drones, and the problem of certifying geomatic surveys for updating the Geodatabase of the Emilia-Romagna Region (conducted in collaboration with the Cartographic service of the Emilia-Romagna Region) were also the subject of research. High performance mobile survey vehicles (MMS) were the subject of work in the thesis path, and in some subsequent contributions, with particular attention to the accuracy in the reconstruction of trajectory and attitude of the vehicle depending on the integration methods between GNSS and inertial data.
The research topics tackled over the past year have mainly focused on three themes: the first concerns the behaviour of low-cost GNSS sensors when used in RTK mode in critical environments from the point of view of satellite reception, with particular reference to the flight of drones in the vicinity of structures and infrastructures under inspection. The second theme concerns the framing and analysis of time series of Persistent Scatterers InSAR, i.e. displacement data obtained from satellite radar interferometry, in particular calculated on the basis of Sentinel observations using SNAP-STAMPS software. The third research topic followed in the recent period concerns the definition of height profiles in remote territories by means of PPP calculation in kinematic mode, implemented in the new JPL GipsyX software, with particular reference to GPS data recorded in Antarctic territory during the ITASE98-99 traverse.
