Research On Dynamic Grasping System Of Robot Based On Machine Vision

With the rapid development of science and technology,robot technology based on vision is widely used in automated production lines.Under the guidance of vision,robots can flexibly and adaptively grab and sort the workpiece on the production line,which has become the demand of more and more enterprises.In order to improve the efficiency and accuracy of production line sorting and effectively reduce labor costs,this topic developed a set of robot dynamic grasping system based on vision,which can effectively solve the problem of dynamic adaptive grasping in the production line,and has a certain positive significance to promote industrial automation and intelligence.In this paper,the robot system,vision system and transmission system related hardware selection and design,and can support a variety of robot end-effectors.The internal and external parameters of the industrial camera are calibrated and the distortion is eliminated,the coordinate system calibration among the robot coordinate system,the camera coordinate system and the conveyor belt coordinate system is completed,and the transformation relationship between the coordinate systems is established,which lays a foundation for the next accurate graspThe image collected by the camera was preprocessed,edge detection,contour extraction,etc.,to effectively identify the contour of the target workpiece,and to complete the extraction of the minimum envelopment matrix of the contour,so as to obtain the centroid position and attitude Angle of each target workpiece.An image de-duplication algorithm based on time and position was designed to effectively remove the repeated objects in multi-frame images and prevent the same object from being passed into the capture queue for many times.The feasibility of the algorithm was verified by experimentsThe dynamic speed detection algorithm of the conveyor belt is designed,which can obtain the position information of the workpiece on the conveyor belt in real time.The dynamic tracking strategy and the workpiece pose adaptive grasping algorithm were designed to realize the robot to dynamically track the workpiece and automatically adjust the attitude of the end-effector to grasp.In the grasping process,the robot used the S-curve acceleration and deceleration algorithm and the gate shape path to optimize the motion trajectory,so as to ensure the stability of the movement.The hardware and software test platform of the robot dynamic grasping system was built to grasp the workpiece moving on the conveyor belt.The experimental results show that the robot can accurately and stably track and grasp the workpiece dynamically,which can meet the design requirements.