Research On Motion Control System Of Fruit Harvesting Manipulator Based On STM-32

Because of the low efficiency of traditional agricultural production,the high cost of labor and the greater impact of natural risks,since the 21 st century,the intellectualization of agricultural equipment has become a solution to the problem of modern agricultural production.As a typical representative of intelligent agricultural equipment,fruit harvesting robot can work in hard environment with high efficiency,so it is widely used in agricultural economic production.The work of this paper is divided into two parts: manipulator motion control and vision recognition.Motion control is to achieve the harvest of fruit.A kind of manipulator motion control system based on STM-32F103 processor is designed,and the micro controller is used as the core to complete the design of manipulator motion control module.Using D-H parameter method to study the forward kinematics and inverse kinematics of the multi joint manipulator,the kinematics simulation model of the manipulator is established,the kinematics analysis and kinematics pose solution are carried out,the coordinates of the manipulator in space and the joint inertia of the manipulator link are analyzed,and the conversion relationship between the two is obtained.According to this relationship,the three-dimensional manipulator is carried out The correctness of the forward and inverse kinematics solutions is verified by the simulation of job trajectory tracking.A motion control module of PC + stm-32 is designed to track the motion control trajectory of the manipulator.After the improved algorithm runs in the controller,it only takes 0.2s to get the calculation result,and the motion control error is within 1%.After the test of the trajectory positioning accuracy of the manipulator,the repeated positioning accuracy is calculated to be 0.0195 mm,and the end of the manipulator actually reaches the target point and is set The average error is 2.5985 mm,which is within the allowable range of the experiment.The camera needs to be calibrated visually,and then the information of the target fruit is collected.Image processing technology is used to collect the information of the target fruit.After the acquired image,the image enhancement,edge extraction,feature recognition and other processes are carried out.There are two methods of target recognition: Based on color and based on shape.In this paper,indirect calibration is used,and the error is about 0.418237 pixel by the result of stereo calibration.The calibration error is very small,and the calibration accuracy of the vision system using binocular vision technology is relatively high.The binocular camera is used to obtain the information of the target fruit and analyze its position coordinates in the space,so as to control the manipulator to complete the operation.After the kinematic analysis of the manipulator,a simple experiment platform of the manipulator motion control system is built by using the theory of binocular vision.The experiment platform is composed of the manipulator body,the manipulator controller,the image acquisition binocular camera and the image processing computer.Finally,by integrating the vision system and the motion control system,a motion control system of fruit harvesting manipulator is formed.The experimental results show that when the camera is less than 40 cm from the target fruit,the absolute error is about 2.5 mm,and the measurement error can be controlled at about 1%,so the measurement accuracy is high.The analysis of the results shows that the scheme designed in this paper is feasible.In order to achieve high intelligence,it is very important to embed the vision subsystem into the robot arm to form a motion control system of fruit harvesting robot arm.The experimental results show that the motion control system of fruit harvesting robot arm in this paper can achieve fruit harvesting.