Research On Control System Of Economical Mounter Based On ARM+FPGA

The SMT machine is an important component of the entire assembly line.Its performance directly determines the quality of the mounted,and the control system is its core part.This article takes the economical arch mounter for small and medium-sized enterprises as the control object,combined with its specific mechanical structure and mounting process requirements,from the point of reduce equipment cost and improving efficiency,a ARM+FPGA is built the core SMT machine control system.In the control system,the ARM side is responsible for task scheduling and distribution mainly based on logic control,and the FPGA side is responsible for tasks such as multi-axis position control and image processing,which are mainly based on a large number of fast data processing.The thesis conducts detailed research from three aspects:motion control system design,machine vision chip position detection,and FPGA-based motor interpolation acceleration and deceleration interpolation algorithms.Firstly,according to the design requirements,and the analysis of the mechanical structure of the SMT machine and its motion control scheme,the control system scheme of ARM+FPGA is determined.According to the preliminary analysis of the control system algorithm,this thesis chooses Altera Cyclone V series 5CSEMA5F31C6N as the core of the controller.In order to speed up the process of system design and experimental research,and combined with the resources required by the control system solution,the DE1-SoC control board of Youjing Technology is used as the platform.Linux+RT extension is used to form a real-time have operated system so as to meet the real-time requirements of the OS.From an economic perspective,on the basis of meet the design requirements,the drive system and image acquisition unit are also designed.Then,to correct the position and angle of components with machine vision technology,a detection method based on Hu moment and progressive Hough is proposed.The median filtering method is used to eliminate image noise,the maximum variance method(OSTU)is used for binary the image.By scanning and counting the area of each enclosed area in the image,the area of the body is set the parameter S.By removing the enclosed area that the area is smaller than S,the pins are removed and the body of the component is obtained.The contour of the component body is extracted by eight neighborhood method tracking.The position and angle of the component are detected by the Hu moment and the progressive Hough algorithm.The algorithm is implemented by pipeline mode on the FPGA side and is validated.Then,in order to improve the control performance of the server system,and to avoid stepper motor out of step or overshoot,based on the analysis of trapezoidal and S-type acceleration and deceleration algorithms,S-type acceleration and deceleration algorithm is selected as the control method of the stepping system,and the S-type acceleration and deceleration algorithm is analyzed and optimized to improve the performance of the motor.Verilog language is used to realize the S-shaped acceleration and deceleration curve under FPGA.ModelSim simulation results show that the algorithm can effectively improve the stability of system operation.Finally,in order to facilitate the debug of the control system,on the premise of analyzed the requirements of the economical mounter control system and the requirements of the functional modules,a simple debug interface is designed based on Qt,to realize the human-machine information interaction of the mounter control system.Tests show that the functions of the designed economical SMT machine control system have reached the basic requirements,which can effectively reduce the cost of the control system and provide a strong support for the subsequent research of the economical SMT machine control system.