Visualization of Space Occupancy Uncertainty in a 3D Voxel-based Urban Model

Abstract. In a 3D voxel-based digital urban model - or digital urban twin, the semantics of each cell is typically visualized using RGB values to give a realistic impression. However, due to measurement uncertainties or incompleteness, some voxel occupancy might be less likely than others. This paper suggests computing an occupancy grid map using an inverse sensor model and embodying the obtained occupancy probabilities into the urban digital twin.
This article then presents visualization of additional information, in this case the uncertainty in space occupancy, in a 3D voxel-based urban model. To this end, a concept is proposed, in which static visual variables are complemented by dynamics. It suggests to utilize the visual variables transparency, size, and color, supplemented by animation. The main idea is to communicate occupancy uncertainty information and enhance its perception via visual variables while visually inspecting a digital twin in an animation mode, observing the occupancy uncertainty without losing the perception of the urban environment.