@article{WeltinLuckeJooste2024, author = {Weltin, Michael and Lucke, Dominik and Jooste, Johannes}, title = {Automatic content creation system for augmented reality maintenance applications for legacy machines}, journal = {Procedia CIRP}, volume = {120}, number = {56th CIRP Conference on Manufacturing Systems, CIRP CMS '23, South Africa}, issn = {2212-8271}, doi = {10.1016/j.procir.2023.09.070}, institution = {ESB Business School}, pages = {750 -- 755}, year = {2024}, abstract = {Augmented reality (AR) applications have great potential to assist maintenance workers in their operations. However, creating AR solutions is time-consuming and laborious, which limits its widespread adoption in the industry. It therefore often happens that even with the latest generation machines, instead of an AR solution, the user only receives an electronic manual for the equipment operation and maintenance. This is commonplace with legacy machines. For this reason, solutions are required that simplify the creation of such AR solutions. This paper presents an approach using an electronic manual as a basis to create fast and cost-effective AR solutions for maintenance. As part of the approach, an application was developed to automatically identify and subdivide the chapters of electronic manuals via the bookmarks in the table of contents. The contents are then automatically uploaded to a central server and indexed with a suitable marker to make the data retrievable. The prepared content can then be accessed for creating context-related AR instructions via the marker. The application is characterized by the fact that no developers or experts are required to prepare the information. In addition to complying with common design criteria, the clear presentation of the contents and the intuitive use of the system offer added value for the performance of maintenance tasks. Together, these two elements form a novel way to retrofit legacy machines with AR maintenance instructions. The practical validation of the system took place in a factory environment. For this purpose, the content was created for a filter change on a CNC milling machine. The results show that inexperienced users can extract appropriate content with the software application. Furthermore, it is shown that maintenance workers, can access the content with an AR application developed for the Microsoft HoloLens 2 and complete simple tasks provided in the manufacturer's electronic manual.}, language = {en} }