Urban digital twins should reflect the urban infrastructure semantically and geometrically as accurately as possible, but also in a timely manner. Therefore, regular updating of existing existing point clouds is necessary in order to do justice to infrastructural changes, e.g. in the context of construction measures. For this purpose, cities often use terrestrial laser scanning or mobile ground-based laser scanning (car, bicycle, backpack), as this allows small areas to be re-measured efficiently. These "new" point clouds must then be efficiently integrated into the existing point clouds so that consistent geoinformation is created. This presentation discusses solutions in this regard and describes a workflow that automatically registers point clouds in as-built point clouds despite many outliers and disturbance points, depending on the geometric richness of the urban development.