Research On Automatic Assembly Of Small Oil Pump Based On Machine Vision

In the production and manufacture of small and medium-sized parts such as small oil pumps,the parts assembled are usually between several centimeters and tens of centimeters,and the assembly accuracy is relatively high.At present,for the assembly of such parts,most enterprises(especially small and medium-sized enterprises)rely on manual work to complete,the work efficiency is low,the workers’ proficiency is high,and it is difficult to guarantee the product consistency of the same batch of manual assembly.With the integration of resources and the centralization of product production,the requirement of product quantity,assembly quality and stability has been improved,and automatic assembly has become a trend.The automatic assembly system based on machine vision has high flexibility,and can adapt to the assembly of many similar parts after simple adjustment.The assembly quality is good and the precision is high.In the automatic assembly system,the classification and positioning of parts are realized mainly by machine vision.After obtaining the shape,position and posture of parts,the information is fed back to the servo system for adjustment.In order to improve the assembly quality and efficiency of small and medium-sized parts manufactured in batches,an automatic assembly system of small-sized rotor oil pump based on machine vision was developed,taking YB100 small-sized rotor oil pump manufactured by an enterprise as the research object.The automatic assembly platform includes the feeding system of the SCARA joint robot,the rectangular coordinate manipulator system,the visual correction system and the upper computer control system.The visual correction system is mainly composed of industrial cameras,light sources and rotary platforms.Before the vision system recognizes and adjusts the target parts,it first needs to calibrate the camera to obtain the camera’s internal parameters.Five different parts are classified by calling GMM Gauss mixture model of HALCON vision library.Then the NCC template matching method is used to identify the position of the target parts and obtain the pose of the parts.Then the angle information is fed back to the PLC,which controls the rotating platform to correct it.Finally,the right coordinate assembly manipulator is controlled to grasp and assemble the parts.Experiments show that the automatic assembly platform can improve the consistency of assembly products,reduce the technical requirements and labor intensity of workers,and create conditions for the industrial application of the device.