Research On Spacecraft Embedded Structure Assembly Defect Detection Technology Based On Visual Image

Spacecraft embedded structure is widely used in the field of aerospace engineering and is often used as the main frame bearing structure of spacecraft.Honeycomb sandwich is a typical form of spacecraft embedded structure.Its core layer is used to maintain the external shape of spacecraft,realize the connection of different cabin structures and load transfer,and meet the requirements of spacecraft for equipment installation and load bearing.With the increasing requirement of spacecraft structure lightweight and complexity,high requirements are put forward for the assembly accuracy of embedded parts.Traditional manual detection methods cannot meet the requirements of mass,high precision and high efficiency of embedded parts detection.In this paper,the honeycomb sandwich embedded structure of spacecraft was studied,and the automatic assembly detection of honeycomb sandwich embedded panel was taken as the target.A binocular stereo vision detection platform was built for the detection of embedded parts assembly.Three-dimensional point cloud algorithm and two-dimensional image algorithm were used to solve the problems of embedded parts localization and identification.By analyzing the main assembly defect characteristics of embedded parts such as multiple assembly,missing assembly and misassembly,the effectiveness of assembly defect detection algorithm of spacecraft embedded structure was studied,and the feasibility of automatic detection technology of honeycomb sandwich embedded structure assembly was explored.Based on visual image detection technology,this paper innovatively proposed a collaborative processing method of three-dimensional point cloud and two-dimensional image for assembly defect detection,used multi-fusion filtering algorithm to remove noise and invalid points from honeycomb point cloud,and used FCM clustering algorithm optimized by whale algorithm for image segmentation.A fast target contour matching method based on Hu moment and hausdorff distance is designed.The whole algorithm is verified by building a honeycomb sandwich embedded structure detection platform.Experimental results show that the defect detection algorithm proposed in this paper can obtain a good detection effect for the embedded component assembly defect detection scenario of spacecraft honeycomb sandwich embedded structure.The research results of this paper fill the gaps in the automatic visual detection of honeycomb sandwich embedded structures in China and provide a basis for the automatic detection of assembly defects of spacecraft embedded structures.