The Development Of Visual Aided Grasping System For Hanging Piece Robot

The textile manufacturing industry wants to develop in the direction of high efficiency,high quality and low cost.The premise is to increase the manufacturing process of the textile process and reduce the amount of labor.The birth of a clothing hanging system has greatly improved the efficiency and quality of clothing production,but there are still many non-intelligent processes in it.In the hanging piece workstation,the work of hanging the pieces of clothes on the hangers needs to be completed by workers.With the extension of working hours,the work efficiency of workers is greatly reduced,and the error rate is also increasing.In order to solve the problem that the hanging piece workstation is difficult to be intelligent,and speed up the automation process of the textile industry,this paper develops the visual aided grasping system of the hanging piece robot.The following major work has been done:(1)Establish reference coordinate system to describe the imaging process of the camera.The principles and models of binocular vision are discussed,and the more applicable general binocular vision model is selected as the binocular vision model of this article.A visual aided hanging piece robot hardware system was designed and built.(2)Using a non-traditional camera parameter solving method.Separate the camera distortion parameters from the camera's internal and external parameters,and solve these parameters separately.Compared with traditional algorithms,the algorithm is simple and easy to understand and is more suitable for practical use.Using the method proposed in the paper to experiment,the accuracy and validity of the algorithm are proved.(3)This paper presents a method for finding the minimum circumscribed rectangle of the matching point set to locate the image coordinates of the finger clip of coat hanger.According to the local feature matching algorithm,the image matching point set of hanging hanger is obtained.The convex hull of matching point set is calculated using Graham algorithm,and the image coordinates of the finger clip are calculated by combining the distance ratio between the actual finger clips.According to the previous algorithm,the algorithm flow from the finger clips image coordinates to the three-dimensional coordinates of the finger clips is deduced,and the purpose of positioning the three-dimensional coordinates of the finger clips is achieved.(4)Analyze the requirement of visual aided hanging piece robot control software,use C++ language to develop control software,design and implement each module of the software,and verify the feasibility of the visual aided grasping system of the hanging piece robot through experiments.Summarizes the current research work completed and looksforward to the future development trend of the project.There are 43 figure,7 tables and 63 references in this paper.