Surface Representation and Area-wise Deformation Analysis

Deformation analyses of man-made objects and natural structures are one of the core competencies of engineering geodesy. The deformation analysis includes the metrological sampling of the potentially deformed object, the evaluation of its geometric changes and the interpretation of the results. In general, infrastructure facilities like bridges, water dams or tunnels are of special interest since their operational reliability needs to be ensured at all times. Damages that are detected too late lead to serious human and economic losses. There is a high probability that the demand of geodetic deformation analyses will increasingly expand in the near future. For example, in several parts of Europe the operational reliability of the bridge constructions in the trunk road system suffers from the increase of commercial transport on highways and from the increase of axle load limits. Increasing the lifetime of such constructions by geodetic deformation monitoring can save a large amount of money and/or sets up efficient repair works. In this context, geodetic deformation analysis demands to detect deformations as early as possible to prevent the building from damages and, thus, to reduce the repair costs and to expand the lifetime. So far in deformation analyses, the object is discretized by individual signalized points. The preselection of an appropriate number and position of these individually chosen points as well as the respective measurement dimension demands interdisciplinary collaboration to obtain knowledge about the expected deformations. Contrarily, area-wise deformation analysis does not rely on the decisions on where to place individual monitored points and, therefore, allows for the detection of non-expected deformations. This research unit aims at the development of a holistic multi-scale area-wise mathematical-statistical deformation analysis based on point clouds captured by terrestrial laser scanners. This raises the needs for new research methods and research findings. These methods relate to challenges that are addressed in single projects and span the whole data processing chain comprising the setup of a complete instrumental error model for terrestrial laser scanners, the description and quantification of the measurement and model uncertainty as well as the objects surface representation and the estimation of deformations from its changes. This project refers to the latter topic, by investigating a functional and a stochastic approach to model point clouds and subsequently derive deformations. Within the functional approach, it is explored whether a refined geometric approximation of the objects surface provides an added value in terms of localisation of deformations as well as of their estimation. The second approach researches the benefit of a stochastic surface representation in terms of uncertainty and correctness of the deformations estimation.