Aviation equipment measurement and assembly analysis method based on robotic system.

• Feature extraction method can extract required data from the raw scanning data. • Derive the optimal position of leapfrog balls and improve calibration accuracy. • Use SVM to determine the parameters of edge-determination method. • Complex components can be measured and analyzed. • The proposed method can significantly reduce time consumption. In the manufacturing industry of aviation equipment (e.g., ballistic missile, rocket, unmanned aerial vehicle (UAV), and airplane), the traditional assembly quality inspection methods mainly rely on manual operation of single- or multi-stations laser trackers, but they are powerless to measure and evaluate complex components on the aviation equipment. This paper introduces a method for automatically detecting and evaluating assembly accuracy: first, the global accurate calibration is performed automatically by deriving the optimized positioning of leapfrog balls; after selecting the aviation equipment type to determine the measurement tasks, each side of the aviation equipment is equipped with a robot and an accompanying FARO scanning arms that can move together on the U-shaped guide rail to the measurement station where the measurement task is to be performed; at each measurement station, the robot drives the movement of the FARO scanning arm through a linkage mechanism to perform scanning; after the whole measurement processes are completed, data from multiple measurement stations is integrated into the same coordinate system for calculation and assembly quality analysis. The algorithm module consists of path planning function for scanning various parts, feature extraction function for various geometric elements, and assembly quality evaluation function. Compared with laser tracker method, the proposed method can effectively and accurately measure and analyze basic measurement tasks: the maximum differences of length, concentricity, and perpendicularity are all less than 0.3%; while the time consumption can be significantly reduced to less than 5% of that of the laser tracker method. Moreover, the proposed method can also achieve the automatic measurement and analysis of complex positioning components that cannot be achieved by laser trackers, e.g., center of positioning hole, position deviation of the bump, and azimuth angle between the positioning surfaces. [ABSTRACT FROM AUTHOR]