A method for improving absolute positioning accuracy of industrial robot in entire workspace domain based on stereo vision

The absolute positioning accuracy of industrial robots is low. At present, kinematics calibration based on the motion loop method is often used to improve the absolute positioning accuracy of the manipulator. However, the size of the calibration space limits the absolute positioning accuracy of the robot's entire workspace. To solve this problem, a method is proposed to improve the absolute positioning accuracy of the robotic arm in the entire workspace domain using the fusion of stereo vision and two- dimensional checkerboard targets. First, the industrial robot is rotated on a single axis and a stereo camera is used to record the single-axis rotation trajectory. The Rodrigues rotation is then used to perform three-dimensional curve fitting to obtain the rotation axis of the industrial robot. Second, a joint coordinate system is established according to the D-H rule and the kinematic parameters of the industrial robot are calculated according to the relative positional relationship between the joint coordinate systems. Third, the reduction ratio of the joint motor is calibrated by analyzing the relationship between the actual rotation angle of the joint and the output value of the encoder. Finally, the proposed method is verified by experiments and the results are compared with those of other accuracy improvement methods. The experimental results show that the absolute positioning accuracy of the manipulator is improved by 67% after calibration using the method in this study and that the effect is the same inside and outside the calibration space, which can improve the absolute positioning accuracy of the industrial robot in the entire working space domain. © 2023 Chinese Academy of Sciences. All rights reserved.