A supplier in the aviation industry applies its seals to screws for aircraft carriers and fuselage sections by hand using a sealant applicator. This can lead to deviations in the shape of the seal and the quantity and position of the sealant. The use of Augmented Reality is intended to optimize this process and thus improve seal quality and durability.
In a joint project with BIBA – Bremer Institut für Produktion und Logistik GmbH, we develop a mobile documentation and inspection system for the application and evaluation of sealants with pseudometric freeform surfaces.
The use of deep learning algorithms results in an inspection system for soft seals, which is continuously re-learning, offers a high reliability and could also be used in various other applications.
The system is mobile and is worn on the body in the form of data glasses in direct human-technology interaction. Position and movement sensors on the sealant applicator as well as an interface for wireless data transfer enable precise process documentation. The data obtained in this way can be used for process optimization, predictive process control and automatic recognition of test points for the camera system. Thus, the process is evaluated in real time during the execution of the work task and, if necessary, corrected by appropriate feedback.