Environment and object design for 3D simulation in context of commercial vehicles

Modeling of objects and environment for simulation in the context of commercial vehicles can be obtained by measuring manually and using those data to create the model in CAD software. This modeling method is time-consuming against the state-of-the-art modeling method using triangulation which is faster and easy to implement. There exist several methods for the generation of triangle meshes. However, the model produced by these approaches is not editable for various simulation scenarios. Therefore, in this paper, a new approach by using the triangulation method is developed which can be used as an element for Vrep simulation that is capable to overcome the challenges (editable and faster) presented by the previous approaches. In this paper, different mesh reconstruction algorithms such as Poisson Surface Reconstruction and Ball pivoting algorithm were applied and their output meshes were assessed. The Poisson surface reconstruction algorithm proved to be quite robust against the Ball pivoting algorithm and is therefore described step by step. Additionally, object separation techniques were applied for grouping different parts. The output meshes using the triangulation method using from both state-of-the-art and the novel approach presented in this paper are compared in terms of size and dimensional accuracy. A pillar height, an entrance area, and road width were measured and the dimensional accuracy was found to be 98.64% for both approaches where the proposed approach has the benefit of being editable.