Research On Control Technology Of Nuclear Rods Assambly Robot Based On Visual Servoing

As the nuclear fuel rod with high radiation and nuclear fuel rod assembly process great harm to the human body,the nuclear fuel rod assembly needs to be automated,and the use of industrial robots instead of artificial assembly has become a trend.Taking into account the slender structure of nuclear fuel rods and the need for stable accurate assembly,this paper,based on the visual servo system,improves the assembly efficiency of nuclear fuel rods.To grasp the nuclear fuel rod from incoming material position to the camera field of view,it is necessary to perform kinematics analysis on the 6-DOF robot.In this paper,the D-H coordinate system is used to analyze the forword kinematics and inverse kinematics of the robot,and the working space of the robot is drawn based on the forword kinematics.According to the simulation curve of trajectory planning of joint space and Cartesian space by Matlab simulation model,the optimal joint interpolation algorithm is selected to ensure the continuous acceleration curve of each joint.To realize the automatic assembly of the nuclear fuel rod,an image-based uncalibrated visual servo method is used to guide the movement of the nuclear fuel rod.This method uses the orthogonal binocular cameras fixed on the workbench and selects the feature points and the feature lines of the image space as the image feature,and the image processing platform is established by MFC,the feature information of the image space was extracted.The image Jacobian matrix is used to approximate the hand-eye relationship by using the adaptive kalman filtering method,and maps the image plane image features to the robot workspace.The image-based uncalibrated visual servo simulation platform is built by using the Motoman_MH5LS robot model,the industrial camera model and the Simulink model,and the 3-DOF and 6-DOF positioning simulation experiment of 6-DOF robot 's uncalibrated visual servo system was carried out.To realize the visual servo system based on dynamics,the robot performs positioning and trajectory tracking analysis.The robot dynamic analysis is set up based on Lagrange dynamic equations.And the PD control algorithm and the PD + feedforward control algorithm are designed based on the dynamic model,and the simulation is carried out by combining with Matlab / Simulink and Adams software.The simulation curves are used to compare the fixed point control and trajectory tracking control performance of each algorithm.The simulation results show that the PD + feedforward control algorithm can realize trajectory tracking of the robot joint.The influence of the PD+ feedforward control strategy on the visual servo system is simulated by combining the image-based uncalibrated visual servo system with the robot dynamics control.