A two-step self-calibration method based on portable calibration devices is proposed in this paper. In the first step, the distance errors in a large-range workspace is utilized to calibrate the manipulator. In the second step, the position errors is utilized to calibrate an external world frame to describe all the coordinate systems in a robot cell and further enhance the calibration results in the first step. For each step, a linear and simplified calibration model is then formulated by employing the local POE formula and introducing the position adjoint conversion matrix. Based on the calibration model, portable and cost-effective self-calibration devices are designed, which consist of a spherical center measuring device, a movable ball bar and Tri-ball plate. The calibration devices presented can obtain distance errors in a large-range workspace and position errors in a local workspace of a manipulator, which enhance the reliability of the calibration results and set an external reference frame. Finally, an accurate kinematic model with respect to the external user-defined frame is given, which enables the off-line programming to be more accurate and effective. The simulation results demonstrate that the proposed two-step self-calibration algorithm is effective and robust.
CITATION STYLE
Gu, L., Yang, G., Fang, Z., Shen, W., Zheng, T., Chen, C., & Zhang, C. (2019). A two-step self-calibration method with portable measurement devices for industrial robots based on POE formula. In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) (Vol. 11740 LNAI, pp. 715–727). Springer Verlag. https://doi.org/10.1007/978-3-030-27526-6_63
Mendeley helps you to discover research relevant for your work.