Easy Pose-Error Calibration for Articulated Serial Robot Based on Three-Closed-Loop Transformations

Most of the existing robot calibration methods have high technical requirements for the installation of end tools and detection sensors as well as other calibration equipments involved, which makes the calibration cumbersome and expensive. This article proposes a novel and easy pose-error calibration method for articulated serial robots based on three-closed-loop transformations. First, the principle of the three-closed-loop transformations of the coordinate systems is introduced, where the loops can be mutually verified to improve the accuracy of the calibration. Then, a hand-eye calibration method that can directly obtain the coordinate conversion between the camera and the tool without need to consider the installation or transformation error of the tool at the robot end is proposed. Furthermore, the entire calibration process is economical without using any expensive special instruments, and the use conditions of the calibration equipments are loose. Finally, experimental results show that the pose error of the robot can be precisely detected and verified by different closed loops.