Borescopes are a widely used technology for optically inspecting machinery from the inside. Recent solutions include enhanced metrology techniques for acquiring image data in high resolution as well as 3D surface information. These already foster more detailed health and damage inspections. However, both inspection and diagnosis strongly rely on the expertise of professionals. Despite their strong coupling, an increasing separation between these tasks is trending. Furthermore, many regions often cannot be inspected properly due to limitations in accessibility and hardware, which is particularly true within toroidal cavities. As a consequence, significant demands regarding emerging digitalization and automatization such as reproducibility, completeness or the steady and high quality of results cannot be fully satisfied. These drawbacks can be faced with a defined probe guidance while providing adjustability towards sensor characteristics. In this publication a novel approach with great automation capability is presented, facing the contactless and complete inspection of toroidal cavities, requiring no disassembly. The proposed solution is related to continuum robotics while based on tube forming technology coupled with a path planning strategy.
Die speziell zur Inspektion von Kavitäten eingesetzte industrielle Endoskopie ist im Gegensatz zu anderen zerstörungsfreien Prüfverfahren bisher wenig automatisiert. Dies liegt größtenteils an der anspruchsvollen Handhabung der Geräte innerhalb schwer zugänglicher Bereiche sowie der komplexen Auswertung der aufgenommen Messdaten aufgrund hochvarianter Aufnahmebedingungen. In diesem Beitrag wird ein neuartiger Kontinuumsroboter vorgestellt, der die kontaktfreie Führung des Sondenkopfs auf einer Kreisbahn erlaubt und so reproduzierbarere und bessere Sichtbedingungen schafft. Zusätzlich wird ein Konzept zur automatisierten Messdatenauswertung basierend auf Deep Learning vorgestellt.
In contrast to other non-destructive testing methods, industrial endoscopy, which is used specifically for inspecting cavities, has so far been little automated. This is largely due to the demanding handling of the devices within areas that are difficult to access as well as the complex evaluation of the recorded measurement data due to highly variant recording conditions. In this paper, a novel continuum robot is presented that enables non-contact guidance of the probe head on a circular path, creating more reproducible and improved viewing conditions. Additionally, a concept for automated measurement data analysis based on Deep Learning is presented.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.