BRIX | AR- and AI-supported analysis of damage progression in structural inspection

The support of inspection tasks for bridge construction projects through digitalisation has been the focus of many research projects over the last few years. As part of this project, available technologies and our own preliminary work will be combined to provide the most practical support possible for structural inspectors using augmented reality (AR) and artificial intelligence (AI). The focus is on visual inspection within the framework of DIN 1076 in conjunction with the digital twin concept. The methods developed in this project will automatically record visible damage to the structure, evaluate it in interaction with the user, and localise it in the digital twin. The latter facilitates the localisation and historical analysis of damages. The following challenges present themselves in this project:

• Modular concept of a digital twin to link historical building information such as inspection reports, digital information and data recorded on site.
• Precise localisation of the mobile AR system to localise identified damage in the digital twin and display it during subsequent inspections.
• AI procedures to facilitate damage assessment, which can be carried out on mobile AR systems in real time, as well as decision support for maintenance measures.
• Development and implementation of a concept for damage prediction based on, among other things, chronological damage patterns and data from depth sensors.
• Improvement of the AI process by incorporating new evaluation results in the course of the building inspection.
• Compliance with information security and data protection requirements during the design and realisation of a demonstrator.