When manually processing complex contours or surfaces, there is a risk of incomplete or unnecessary multiple processing. The support of such processes with an augmented reality (AR) application will be presented in this work using the example of manual deburring. Firstly, the positions of the component and tool are determined based on camera recordings. For this purpose, approaches for 6 Degrees of Freedom (DoF) pose estimation will be utilized. Secondly, the correct and complete processing is concluded by further analyzing the extracted data. For example, the distance of the objects to each other. In the third step, the processing result is visualized to aid the worker. Finally, the system is validated through a study. The results show that with the help of image targets a robust recognition of the objects at a normal execution speed and thus process support can be realized.
Science Journal 