3D reconstruction of as-built industrial instrumentation models from laser-scan data and a 3D CAD database based on prior knowledge

In the operation and maintenance phase of industrial plants, it is important to ensure that the as-built condition of every piece of installed instrumentation, and any changes that may subsequently occur, are properly recorded and adjusted in interpretive documents. However, the manual creation of industrial instrumentation models from an acquired set of three-dimensional (3D) point clouds is difficult. This paper proposes an approach for the reconstruction of 3D models of as-built industrial instrumentation from terrestrial laser-scan data and a 3D computer-aided design (CAD) database based on prior knowledge. An experimental evaluation of an actual industrial plant demonstrates that incorporating prior knowledge into the reconstruction process greatly reduces the manual interaction required, facilitating the process of reconstructing 3D models of as-built industrial instrumentation from unstructured (raw) laser-scan data. The proposed approach could be successfully utilized to assist in the reconstruction of 3D models of as-built industrial instrumentation during the operation and maintenance phase of an industrial plant.