Design Of A Six-degree-of-freedom Mechanical Arm Sorting System Based On Monocular Vision

In recent years,people have put forward higher requirements for industrial production efficiency.Industrial intelligentization,flexible production,and integrated sorting systems are bound to be the trend of smart factories in the future.The traditional sorting system is based on fixed sorting method.Once the position of small workpiece changes,the sorting will fail,and it is difficult to sort according to specific requirements.Machine vision technology has been developing faster and faster in recent years.It has become the main research direction of many scholars to add traditional sorting system into machine vision technology.Machine vision-based sorting system is now widely used in food,express delivery,agriculture,manufacturing and other fields.However,for small workpieces,especially for electronic components sorting,there are still some problems,such as poor recognition rate under the influence of illumination and low sorting accuracy of end-effector.Based on the analysis of the problems existing in many kinds of sorting systems and the demand of enterprises for production sorting,a 6-dof mechanical arm sorting system based on monocular vision is designed?Firstly,aiming at the sorting problem of small workpieces to be identified on the production line under complex environment,the sorting method combining monocular vision and six-degree-of-freedom mechanical arm is adopted,and the hardware platform of the sorting system is constructed to constitute the six-degree-of-freedom mechanical arm sorting system based on monocular vision.On the basis of the completion of the hardware construction of the system,D-H modeling was carried out for it,and the 6-dof manipulator was controlled by the reverse solution and forward adjustmentcontrol method proposed,and the sorting action was written and tested.The test results show that the sorting action is more flexible and the grasping precision is improved.Secondly,the above study only performs actions at the end of the sorting process and cannot effectively sort small workpieces according to specific requirements.For small artifacts in the lighting effect and practical part of adhesion and changes in the environment under a variety of problems of the interference,small workpiece based on HSV color recognition algorithm,based on the geometric features of small workpiece shape recognition algorithms,and applications for small workpiece more electronic components for electronic components recognition based on feature fusion algorithm is put forward.The three algorithms are compared with the template matching algorithm applied in the large-scale production line at present,and the test results show that the interference of the recognition effect caused by the light effect and the partial adhesion of small workpieces is reduced,and the correct recognition rate is improved greatly.Finally,this paper develops software through the QT interface,and encapsulates the sorting action program of the six-dof manipulator and the recognition algorithm of small workpieces together to form the human-computer interaction interface of the six-dof manipulator sorting system based on monocular vision.This system can well realize the color identification function,shape identification function,electronic components identification function,positioning function,sorting function and so on of small workpiece,and meet the basic performance requirements of the small workpiece sorting system,and the sorting system has good real-time recognition and sorting,with a wide range of application prospects.