| Industrical robots have been widely used in various fields with the rapid development of modern technologies.The accuracy and reliability of robots are one of the key factors during applications.The actual opertion accuracy is hard to improve with the influence of variant structure and working environment.Error compensation method can be used to solve the above problems.However the existing error compensation methods have some limitations due to the ignoring of the influence of non-geometric error.The kinematic and non-kinematic error models of the six-DOF industrial robots is setup in this paper,with the purpose to analyze the influence of robot accuracy with geometric and non-geometric errors.The extended kalman filter algorithm is developed to identify the kinematic parameters of the robot and validited through numerical simulation.An experimental platform is built to verify the performance of the error compensation method.The main research contents are as follows:A coordinate system conversion method based on feature line coincidence is proposed to solve the problem of the low accuracy of coordinate system conversion in the robot measurement process.The minimum closed sphere is solved based on the robot demonstrator coordinate point cloud data and the measurement system coordinate point cloud data.The conversion matrix is obtained through the feature line extraction and coordinate conversion and verified experimentally.The sources of geometric errors of the robot are analyzed.The kinematic model of IRB-1410 robot is built based on the M-DH method and verified by the MATLAB Robotics Toolbox.The transmission error of the four-bar mechanism in the above robot is analyzed and a transmission error model with the end position error and the geometric parameter error is established.For the non-geometric error sources of the robot,the significant characteristics of the kinematic parameters in the robot’s self heating generation model are analyzed,and a temperature error model is established.The compensation values are added in the kinematic parameters of the robot to realize the temperature error compensation.The relationship between applied load and robot position accuracy is analyzed and verification experimentally.A load error model is established to compensate for the end positioning error.The extended extended kalman filter algorithm is used to identify and compensate the robot kinematic parameters,according to the kinematic geometric error model.An experimental platform is bulit and the performance of the compensation methods is verified.Experimental results show that the repetitive positioning accuracy of the robot is improved 37.2% after compensation.The above robot compensastion methods can realized improvement of robot accuracy effectively,which make the robot operation more realibility. |
PDF Full Text Request