Joint Stiffness Identification Of Industrial Robot Based On Accurate Calibration Of D-H Parameters

Major equipment manufacturing industry has been a hot topic in recent years. In the national five-year plan, the high-end equipment manufacturing industry has been listed as one of the emerging strategic industries. Nowadays, China is developping large-scale construction projects and equipment, such as the large aircraft, aircraft carriers, high-speed rail technology, large-scale offshore engineering, nuclear engineering, etc., they have a huge device, complex structure, high precision, and so difficult to manufacture. Traditional manufacturing methods cannot meet their requirements, which must use precise online measurement and control technology to achieve. The complex issue this article belongs to regarded the process of complex drilling of a large spherical surface as background. The system of on-line measurement and control compensation based on industrial robots and combined photogrammetric system, rails, inclinometer and other instruments was proposed.When industrial robots performing the job, the end often grips some actuators. Because of the impact of the equipment gravity and contact force, the robot end will produce distortion, so that there is a deviation between the actual arrival and expectations pose, which reduce the positioning accuracy of the robot. In this paper, the research on the robot end distortion caused by the load was carried out.1. The structural characteristics and movement ways of KUKA-arc5 industrial robot was introduced. The classical D-H model was used to established the robot kinematics equation by the homogeneous transformation matrix. On the basis of DH model, the forward kinematics and inverse kinematics of robots were analyzed and solved. A brief introduction of the knowledge of differential kinematics of the robot was made. The Jacobi matrix and its transposed matrix force Jacobi matrix was deduced by vector integration method.2. The research on the industrial robot calibration technology was carried on. Factors affecting the robot absolute positioning accuracy were analyzed, where the robot geometry errors were the main factors affecting the absolute positioning accuracy of the robot. Calibration on the robot and getting the actual DH parameters can reduce the calculation error during identification. In this paper, the method based on the robot position error model and least squares were used to calculate the robot's DH parameter errors. The KUKA-arc5 industrial robot was effectively calibrated.3. Based on the traditional concept of stiffness, joint stiffness matrix of the robot and the end Cartesian stiffness matrix was derived in detail. Some nature and significance of the robot Cartesian stiffness matrix were studied. Then Relationship between the robot joint stiffness matrix and the end Cartesian stiffness matrix was built. On this basis, a simple method to identify the robot joint stiffness was proposed.4. In the stiffness identification experiment, static load was applied at the end of the robot, a laser tracker measurement was used to measure the end deformation and the joint stiffness values were calculated by the least squares method. 15 new points were used to verify the reliability of the results.