DICONDE for dent digital twins: Data standardization for aircraft structural inspections

Abstract

Aircraft structural inspections increasingly rely on digital twins to improve the detection, localisation, and assessment of dents and buckles. These processes often integrate multiple data sources, including 2D imagery (e.g. RGB camera data), 3D measurements (e.g. laser-scanner point clouds) and damage assessment algorithm results. However, the resulting datasets are heterogeneous, stored in proprietary formats, and lack consistent contextual information, creating significant challenges for efficient data integration, fusion, and interpretation. This research investigates the suitability of the Digital Imaging and Communication in Non-Destructive Evaluation (DICONDE) standard as a unifying framework to address these challenges in dent and buckle inspections. A data conversion framework is developed to implement the DICONDE standard for 2D image data capture, artificial intelligence (AI)-based damage annotations, point cloud data from laser scanners and STEP (Standard for the Exchange of Product model) data files. The aspects of consistency, traceability and interoperability between datasets before and after DICONDE standardization are assessed. Furthermore, data association and correlation for a multi-inspection data integration and fusion are discussed. This paper contributes to the standardization for the digital data flow for future Maintenance, Repair, and Overhaul (MRO) processes.