| Industrial robots integrated system is an important development direction of modern industrial automation. With the expansion of industrial robots application, the off-line programming is playing a more important role than before. Industrial robots have a high repeat positioning accuracy, but the absolute positioning accuracy is often insufficient to meet the demand for off-line programming applications. So, how to improve the absolute positioning accuracy becomes one of key issues to extend robotic applications. With the analysis of the 6R robot frame and the error sources, this paper tries to develop a new method to improve the absolute accuracy of the robots. A KUKA KR210 robot is used to test. The main research is present as follow:1. Based on the analysis of the coordinate translation between links, the forward kinematics model of the KR210 robot is established by the classical D-H model. Based on Pieper method, the inverse kinematics is solved. Beside, robot positioning error model is derived from the kinematics.2. Base on the analysis of the geometry errors, the robot compliance error model is established. Then, the position error model coupling compliance errors is put forward. In the calibration method, the Levenberg-Marquardt algorithm is used to calculate the optimal solution, and the Descartes method is used to correct the position. Finally, a simulation is designed to test the calibration method.3. Two sampling point selection method are investigated. The first one is the uniform sampling points selection method based on the error similarity. The second is the random sampling points selection method based on the observability.4. A series of experiment is investigated to test the calibration method that proposed before. After calibration the robot’s absolute accuracy is improved to less than 0.3 mm. The feasibility and validity of the strategy is verified. |
PDF Full Text Request