User-Driven Computer-Assisted Reverse Engineering of Editable CAD Assembly Models

Abstract This paper introduces a novel reverse engineering (RE) technique for the reconstruction of editable computer-aided design (CAD) models mechanical parts’ assemblies. The input is point cloud assembly that has been acquired as whole, i.e., without disassembling it prior to its digitization. proposed framework allows parametric CAD model through multi-step and fitting approach. It modular supports various exploitation scenarios depending on available data starting point. also handles incomplete datasets. process starts from roughly sketched parameterized geometries (i.e., 2D sketches, 3D parts, or assemblies) are then used simulated annealing-based algorithm, which minimizes deviation between adapted geometries. coherence maintained by modeler performs geometries’ updates while guaranteeing possibly imposed constraints coherence. optimization leverages two-level filtering able capture manage boundaries inside overall in order allow local interfaces detection. user-driven approach where user decides what most suitable steps sequence operate. tested validated both real scanned clouds as-scanned virtually generated incorporating several artifacts would appear with acquisition devices.