One Side Stitching Seam Detection And Suture Simulation Of 3d Textile Composites Preform Based On Vision

The 3D textile composites have the characteristics of light texture, high strength, good toughness and flexibility. And it will become one of the mainstream direction of new material in High-grade car, aviation and aerospace manufacturing technology in the 21st century.The reinforced fabrics of composites is usually a three-dimensional structure. Therefore, the traditional bilateral suture technique has many restrictions. In order to solve this problem , many countries focused on the single suture technique, and the one side stitching robot has been developed. In addition, in order to improve the flexibility and intelligence of one side stitching robot , the machine vision in textile composites get more and more applications. The seam detection and stitching based on structure light and vision is the key of this paper, and is also the focus and difficulty in one side stitching technique.The study puts emphasis on the one side stitching of glass fiber composites,it studies from the digital image processing and the trajectory tracking technique of seam stitching. The thesis has mainly accomplished the following job in several aspects:(1) The vision tracking stitching system of one side stitching robot has been designed. And a simple visual sensor mathematical model is established, which can achieve the coordinate transformation of the seam effectively.(2) In the image preprocessing, due to the surface of composites reflects seriously and contains a lot of burrs and weaving flowers, image processing methods such as image enhancement, median filtering and open operator of mathematical morphology are used. Thus the clear seam images are gotten.(3) In the post-processing of seam images, the morphological thinning algorithm and the column search method are respectively used for the extraction of seam centerline. Experimental results show that the best seam centerline is extracted by the column search method. The detect search method is adopted to detect feature points of the extracted centerline. According to the deviation calculation formula, the seam deviation is calculated, thus the seam deviation adjustment can be realized.(4) In order to realize the trajectory tracking stitching of seams, first establish the ABB single-sided suture robot coordinate of each joint, analysis the robot by using forward and inverse kinematics. The expression of six joint angles is calculated based on the MATLAB software. Then the seam trajectory tracking is realized by using linear and curve interpolation algorithm. In the trajectory tracking of linear seams and circular seams,experimental results show that the method can track seam trajectory very well.(5) According to the quality requirements of one side stitching,the appropriate stitching parameters and trajectory of slerilized are selected. The line, parabolic and curve seams are stitched by adopting herringbone lines, and satisfactory results are obtained.