Study On Grafting Method And Control System Of High Density Hole Plate Seedlings

This paper proposes a method for taking single-root seedlings on the plate for high-density plug seedlings,and designs a seedling device including a terminal,a manipulator and a control system,and completes the search and positioning of high-density plug seedlings,Shearing,clamping and taking seedlings and other actions.The innovative point of this paper is to focus on the characteristics of high-density plug seedlings.Two seedling terminal and seedling extraction methods based on touch sensing detection and visual depth detection are proposed and designed,which realizes the realization of high-density plug seedlings in the hole The single clamping in the tray provides a basis for the development of automated grafting machines.The main work completed in this research is as follows:(1)Research on image preprocessing method of high-density plug seedlings.Based on the collected images of high-density plug seedlings,this paper studies the features of pixels extracted from the RGB space model and the HSV space model,carries out the pre-processing method for extracting target pixels of plug seedlings,analyzes the formation of connected domains and judges the connected relationship,and reduces the images from the images.Analyze the influence of noise and light intensity on the target contour effect;Secondly,According to the pretreatment method,the edge point of the cotyledon of the plug seedling was judged,and the average maximum error was 0.974mm;finally,the judgment of the direction of the cotyledon of a single seedling was realized,and the average error was about 0.5 degrees.(2)Research on the design of high-density plug seedling collection device and the method of seeking seedlings.According to the characteristics of high-density plug seedlings,a touch-sensitive method of fetching and clipping was first proposed,and a touch-sensitive fetching terminal was designed.Through experimental tests,the seedling positioning and fetching of a single seedling were realized.The maximum error of the Euclidean distance between the theoretical position and the actual position when seeking seedlings is within0.41 mm,and the maximum error of the plane Euclidean distance when cutting seedlings is within 0.17mm;Secondly,a method for taking seedlings from plug seedlings based on visual depth detection is proposed.A terminal mechanism based on visual depth detection is designed.The visual working range of the terminal is within 64 mm to 90 mm from the stem of the plug seedlings,and the distance is obtained by calibration.The experimental results show that the 5th degree fitting polynomial is closest to the scatter change law;the terminal positioning and clipping of the seedlings based on the visual depth detection is realized,the accuracy of the error between the device measurement value and the manual measurement value within 0.1mm is95.5%.(3)Design and research of seedling manipulator and control system.First of all,according to the seedling requirements of this project,the bow-shaped frame and hollow tube of the grafted seedling manipulator were improved and designed,so that the average left and right deviations of the terminal level were reduced to 0.1 degrees and 0.167 degrees,respectively;Secondly,the I2 C communication between the main control machine and the controller and the CAN communication between the main control machine and the manipulator motor are realized.The experiment analyzes the influence of the pull-up resistor on communication,and the result shows that the communication quality is the best when the pull-up resistor is 4.7K and the communication distance is 10 cm,and the communication rate can reach 449.25Kbps;Finally,the interactive commands of I2 C communication and CAN communication are designed to realize the motion control of the manipulator and the seedling terminal by the main control computer.(4)High-density plug seedling taking experiment.Using the seedling terminal and the method,combined with the operation of the control system,the seedling harvesting operation based on touch sensing and the seedling harvesting operation based on visual depth detection were completed.After many experimental tests,the success rates of the two methods reached92.5% and 95% respectively.