Research On Autonomous Planning Of Robot Deburring Slab Based On ROS

With the rapid development of the industrial intelligence in the recent years,robot has been widely applied in industry.But as an important industrial industry in the country,the steel industry still has a larger development space in current level of intelligence.For example,the slab burrs are mainly removed by scraper method and hammer knife method at present,but there are disadvantages of not smooth and incomplete in burrs removal,meanwhile,the slab burrs are unable to be removed when the slab off-track.Therefore,based on ROS platform,an intelligent slab burrs removal scheme based on robots and machine vision is proposed,and the related research is conducted,the specific research content is as follows.(1)The workstation ROS model is constructed.the Soildworks model of robot deburring workstation for slab is built,and it is imported into the ROS system,the model urdf file is converted into an xacro file suitable for simulation by writing code.the xacro file based on the RVIZ visualization interface is used to display the 3D model on the ROS side,the simulation motion controller is added to the robot,so the Move It frame configuration is completed,finally,the Move It and Gazebo are linked to configure the Move It-Gazebo joint simulation model of the workstation.(2)The corner location based on point cloud and ROS node storage of its information.The depth camera module is added to the original Move It-Gazebo model by using the simulation depth camera,and the 3D point cloud reconstruction of the slab is performed,the point cloud coordinates are filtered to obtained the world coordinate data of the slab,and the other data is combined to locate the second corner of the slab.The obtained corner coordinates of the simulated slab are stored in the form of ROS nodes,and the corner positioning of the normal state of the slab and the left and right deviation states are verified respectively,and Mark is designed to display the positioning corner points for inspection.(3)The autonomous planning movement of the robot.The Move It API on the ROS side is used to control the robot,the corner information stored by the node is received and the robot is moved to the corner autonomously based on the RRT-Connect algorithm,according to the principle of Cartesian motion planning algorithm,the robot is designed to autonomously perform linear trajectory planning between the two corners and the deburring work is completed.(4)Matlab assisted simulation research.Matlab and ROS are used to simulate,the ROS side workstation topic information in Matlab is received remotely,and related topics is described respectively;ROS is used to record workstation work package files,the package files on ROS and Matlab are analyzed and studied respectively,finally,Matlab is used to analyze the sixth-axis motion trajectory,and the trajectory function is fitted by the Matlab function fitting tool.the correctness of the trajectory is checked by comparing the deviation of the slab and the slope of the fitting function,and verify the accuracy of the burr removal work.According to the trajectory visualization of the ROS side and the verification of the Matlab trajectory,it is shown that the solution can complete the task of removing burrs and achieve the set work objectives.