Research On Vision Location And Measurement Technology Of Medical Filling And Sealing Robot

With the development of China’s economy and the improvement of medical environment,the domestic pharmaceutical market has maintained a high-speed growth trend.Pharmaceutical industry is related to the production cost and quality of pharmaceutical products.It is also an important industry related to national economy,people’s livelihood and social stability.The high-end intelligent manufacturing equipment represented by medical robots will change the problems of low production efficiency,poor quality and high cost of manual operation,and will also become a new trend of the development of pharmaceutical industry.Pharmaceutical filling robot is an important equipment for filling and sealing of liquid medicine in pharmaceutical intelligent production line,which directly determines the quality and efficiency of pharmaceutical product production.Aiming at the problems of low automation level and poor aseptic in traditional pharmaceutical production line,this paper studies and implements a medical filling robot system based on vision positioning and measurement technology,which realizes automatic and intelligent filling and bottle sealing process.The main work of this paper is as follows.Firstly,the current situation of pharmaceutical market and pharmaceutical industry at home and abroad is introduced,and the application and research of machine vision in pharmaceutical industry are summarized from the application of vision technology in pharmaceutical production lines of major pharmaceutical enterprises at home and abroad.Secondly,aiming at the problems existing in filling and sealing links in pharmaceutical production and the requirement of developing medical filling robot system,the overall scheme of the system is given,and the design of visual system,the selection of devices and the overall scheme of electrical control system are elaborated in detail.Then,aiming at the problems of round surface,non-uniform reflection,virtual boundary,burr on the edge of bottle stopper and bottle cap,texture on the inner surface and adhesion on the edge of bottle mouth,an accurate circle location algorithm based on K-Means++ is studied and implemented.Firstly,the center of bottle mouth,bottle stopper and bottle cap is roughly estimated by barycenter method,and the corresponding outer edge is found by radial scanning;then,the outer edge points are mapped near the center of the circle by three-point circle fitting method to form the center point set;then,the distribution characteristics of the center point set are analyzed by K-Means++ clustering algorithm,and the noise characteristics are eliminated;the real edge mapping points are binary Gauss distribution,and the poles are used.The center points of the circumference of the bottle mouth,bottle stopper and bottle cap are obtained by the method of large likelihood estimation.The experimental results show that the precise circle localization algorithm based on K-Means++ in this paper has higher localization accuracy and better fitting effect than the traditional circle localization algorithm,and can meet the actual needs of the pharmaceutical production line very well.Then,aiming at the phenomenon of image distortion caused by industrial camera lens,camera calibration is studied and image distortion is corrected.A visual measurement method based on space coordinate transformation is studied and implemented.By converting the pixel coordinates of bottle mouth,bottle stopper and bottle cap of the same position into the corresponding robot coordinate system,the precise moving distance of the robot in filling-filling and capping-rolling processes is obtained.The experimental results show that the accuracy of the visual measurement method studied has reached sub-millimeter level.The intelligent guidance of vision system for precise positioning and measurement of medical filling robot is realized.Finally,according to the requirement of filling and sealing process,the software system of medical filling robot is studied and realized.The architecture and functions of the software system are introduced in detail,and the specific implementation scheme of each functional module is analyzed.