We address the problem of automated inspection of rivets and brackets in aircraft fuselage shells with a robot guided 3D-sensor. High variety of such assemblies requires that an inspection automatically compares acquired 3D data to the CAD model representing the assembly. For this purpose, a precise sensor pose with respect to the fuselage shell coordinate system is needed. In this industrial context, the current set-up can deviate from the model by few centimeters which is too much for inspection. We present an automated registration method, which does not need further sensors for tracking. This method reduces the deviation between measured points and the CAD model to few tenth of a millimeter in regions with no assembly defect. Therefore, we find automatically measurement positions for registration which cover all degrees of freedom. The algorithm for registering the 3D points to the CAD model avoids the assignment of points to faces which cannot be reached by the sensor, and parts to be inspected. In addition, local registration can deal with slight deformation of the fuselage shell.
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.