Precision measurement and compensation of kinematic errors for industrial robots using artifact and machine learning

Abstract

Industrial robots are widely used in various areas owing to their greater degrees of freedom (DOFs) and larger operation space compared with traditional frame movement systems involving sliding and rotational stages. However, the geometrical transfer of joint kinematic errors and the relatively weak rigidity of industrial robots compared with frame movement systems decrease their absolute kinematic accuracy, thereby limiting their further application in ultra-precision manufacturing. This imposes a stringent requirement for improving the absolute kinematic accuracy of industrial robots in terms of the position and orientation of the robot arm end. Current measurement and compensation methods for industrial robots either require expensive measuring systems, producing positioning or orientation errors, or offer low measurement accuracy. Herein, a kinematic calibration method for an industrial robot using an artifact with a hybrid spherical and ellipsoid surface is proposed. A system with submicrometric precision for measuring the position and orientation of the robot arm end is developed using laser displacement sensors. Subsequently, a novel kinematic error compensating method involving both a residual learning algorithm and a neural network is proposed to compensate for nonlinear errors. A six-layer recurrent neural network (RNN) is designed to compensate for the kinematic nonlinear errors of a six-DOF industrial robot. The results validate the feasibility of the proposed method for measuring the kinematic errors of industrial robots, and the compensation method based on the RNN improves the accuracy via parameter fitting. Experimental studies show that the measuring system and compensation method can reduce motion errors by more than 30%. The present study provides a feasible and economic approach for measuring and improving the motion accuracy of an industrial robot at the submicrometric measurement level.

Keywords

Industrial robot / Kinematic error / Calibration / Machine learning / Recurrent neural network (RNN)

References

[1] Besset P, Olabi A, Gibaru O (2016) Advanced calibration applied to a collaborative robot. In: 2016 IEEE international power electronics and motion control conference (PEMC), Varna, pp 662–667

[2] Denavit J, Hartenberg RS. A kinematic notation for lower-pair mechanisms based on matrix. J Appl Mech, 1955, 22: 215-221.

[3] Hayati S, Mirmirani M. Improving the absolute positioning accuracy of robot manipulators. J Robotic Syst, 1985, 2(4): 397-413.

[4] Joubair A, Bonev IA. Kinematic calibration of a six-axis serial robot using distance and sphere constraints. Int J Adv Manuf Technol, 2015, 77(1/4): 515-523.

[5] Stone H, Sanderson A, Neuman C (1986) Arm signature identification. In: Proceedings of IEEE international conference on robotics and automation, 7–10 April, San Francisco, pp 41–48. https://doi.org/10.1109/ROBOT.1986.1087664

[6] Zhuang H, Roth ZS, Hamano F. A complete and parametrically continuous kinematic model for robot manipulators. IEEE Trans Robot Autom, 1992, 8(4): 451-463.

[7] Chen IM, Yang GL, Tan CT, et al. Local POE model for robot kinematic calibration. Mech Mach Theory, 2001, 36(11/12): 1215-1239.

[8] Yang X, Wu L, Li J, et al. A minimal kinematic model for serial robot calibration using POE formula. Robot Comput Integr Manuf, 2014, 30(3): 326-334.

[9] Nubiola A, Bonev IA. Absolute robot calibration with a single telescoping ballbar. Precis Eng, 2014, 38(3): 472-480.

[10] Marwan A, Simic M, Imad F. Calibration method for articulated industrial robots. Procedia Comput Sci, 2017, 112: 1601-1610.

[11] Nubiola A, Bonev IA. Absolute calibration of an ABB IRB 1600 robot using a laser tracker. Robot Comput Integr Manuf, 2013, 29(1): 236-245.

[12] Du GL, Zhang P, Li D. Online robot calibration based on hybrid sensors using Kalman filters. Robot Comput Integr Manuf, 2015, 31: 91-100.

[13] Nissler C, Marton ZC (2017) Robot-to-camera calibration: a generic approach using 6D detections. In: The 1st IEEE international conference on robotic computing (IRC), 10–12 April, Taichung, pp 299–302. https://doi.org/10.1109/IRC.2017.66

[14] Joubair A, Zhao LF, Bigras P, et al. Absolute accuracy analysis and improvement of a hybrid 6-DOF medical robot. Industrial Robot, 2015, 42(1): 44-53.

[15] Shibuya Y, Maru N (2009) Control of 6 DOF arm of the humanoid robot by linear visual servoing. In: 2009 IEEE international symposium on industrial electronics, Seoul, pp 1791–1796. https://doi.org/10.1109/ISIE.2009.5218128

[16] Ma G, Wang F, Qu Z et al (2006) A feasible vision-based measurement method for robot orientation error. In: The 1st international symposium on systems and control in aerospace and astronautics, Harbin, 19–21 January, pp 1214–1217

[17] Motta JMST, McMaster RS (2002) Experimental validation of a 3-D vision-based measurement system applied to robot calibration. J Braz Soc Mech Sci 24(3). https://doi.org/10.1590/S0100-73862002000300011

[18] Miseikis J, Glette K, Elle OJ et al (2016) Automatic calibration of a robot manipulator and multi 3D camera system. In: IEEE/SICE international symposium on system integration (SII), Sapporo, pp 735–741. https://doi.org/10.1109/SII.2016.7844087

[19] Oriolo G, Paolillo A, Rosa L, et al. Humanoid odometric localization integrating kinematic, inertial and visual information. Auton Robots, 2016, 40(5): 867-879.

[20] Monica T, Giovanni L, PierLuigi M et al (2003) A closed-loop neuro-parametric methodology for the calibration of a 5 DOF measuring robot. In: Proceedings of IEEE international symposium on computational intelligence in robotics and automation, pp 1482–1487. https://doi.org/10.1109/CIRA.2003.1222216

[21] Liu J, Zhang Y, Li Z. Improving the Positioning Accuracy of a neurosurgical robot system. IEEE/ASME Trans Mechatron, 2007, 12(5): 527-533.

[22] Aoyagi S, Kohama A, Nakata Y et al (2010) Improvement of robot accuracy by calibrating kinematic model using a laser tracking system-compensation of non-geometric errors using neural networks and selection of optimal measuring points using genetic algorithm. In: IEEE/RSJ international conference on intelligent robots and systems, 18–22 October, Taipei, pp 5660–5665. https://doi.org/10.1109/IROS.2010.5652953

[23] Messay T, Chen C, Ordoñez R et al (2011) GPGPU acceleration of a novel calibration method for industrial robots. In: Proceedings of the IEEE national aerospace and electronics conference (NAECON), 20–22 July, Dayton, pp 124–129

[24] Nguyen HN, Zhou J, Kang HJ. A calibration method for enhancing robot accuracy through integration of an extended Kalman filter algorithm and an artificial neural network. Neurocomputing, 2015, 151(3): 996-1005.

[25] Yuan P, Chen D, Wang T, et al. A compensation method based on extreme learning machine to enhance absolute position accuracy for aviation drilling robot. Adv Mech Eng, 2018, 10(3): 1687814018763411.

[26] Chen DD, Wang TM, Yuan PJ, et al. A positional error compensation method for industrial robots combining error similarity and radial basis function neural network. Meas Sci Technol, 2019, 30(12): 125010.

[27] Nguyen HN, Zhou J, Kang HJ. A calibration method for enhancing robot accuracy through integration of an extended Kalman filter algorithm and an artificial neural network. Neurocomputing, 2015, 151: 996-1005.

[28] Leica (2021) Laser tracker systems. https://leica-geosystems.com/products/laser-tracker-systems. Accessed 8 Dec 2021

[29] Faro (2021) Vantage laser trackers. https://www.faro.com/zh-CN/Products/Hardware/Vantage-Laser-Trackers. Accessed 8 Dec 2021

[30] API (2021) Radian laser trackers. https://apimetrology.com/radian/. Accessed 8 Dec 2021

[31] Luo Z, Sturm J. Error analysis. Handb Semidefinite Program, 2000, 27: 163-189.

[32] Tang HY, He ZY, Ma YX, et al. A step identification method for kinematic calibration of a 6-DOF serial robot. Mech Mach Sci, 2017, 408: 1009-1020.

[33] He KM, Zhang XY, Ren SQ et al (2016) Deep residual learning for image recognition. In: IEEE conference on computer vision and pattern recognition (CVPR), Las Vegas, pp 770–778. https://doi.org/10.1109/CVPR.2016.90

[34] Lecun Y, Bengio Y, Hinton G. Deep learning. Nature, 2015, 521(7553): 436-444.

[35] Huang G, Liu Z, Maaten LV et al (2016) Densely connected convolutional networks. In: Proceeding of the conference on computer vision and pattern recognition (CVPR), 21–26 July, Honolulu, pp 4700–4708

Funding

national key r&d program of china(2017YFA0701200)

shanghai science and technology committee innovation grant(19ZR1404600)

fudan university-ciomp joint fund(FC2020-006)