本論文提出一種針對非線性機械手臂的系統鑑別演算法，此演算法能針對高度非線性的機械手臂鑑別出一個最佳化的模型。本論文並非是要鑑別出一組多個線性或非線性子模型，以能在開迴路的狀況下使得這些子模型的輸出、入訊號與機械手臂相同，而是要鑑別出單一個線性的狀態空間模型，使得模擬誤差(存在於被鑑別出的線性模型與高度非線性模型之間)的等效干擾輸入的平方和最小，其中建模誤差為在閉迴路狀況下的狀態以及控制訊號的函數。一旦得到這個最佳化的線性模型，將可以針對具有未知匹配或不匹配干擾的資料採樣非線性系統，以設計線性的狀態觀測器、基於狀態函數觀測器的等效輸入干擾估測器以及可以限制系統輸入訊號大小的模型預測控制器，這些內容都將呈現於本論文中。針對具有未知匹配或不匹配的干擾的連續非線性系統，設計線性狀態估測器、基於輸出迴授的等效干擾估測器(由於此估測器相較於基於狀態函數觀測器的等效輸入干擾估測器較簡單被設計)以及最優化二次數位追蹤器也都將呈現於本論文。除此之外，本論文也將非線性等效輸入干擾估測器以及模型追隨控制，應用於具有未知匹配或不匹配干擾的連續非線性系統，此方法將是上一個方法的可匹敵方法。本論文也提供了一些範例，以說明所提方法的優勢。

An efficient system identification algorithm to achieve the optimal linear state-space model for a class of highly nonlinear robot arm manipulator is proposed in this thesis. Instead of identifying a set of linear/nonlinear sub-models to match with the open-loop input-output data sets, this thesis aims at identifying a single linear state-space model such that the servo-control-oriented squared-sum of the equivalent input disturbance (EID) of the mismatched modelling error (between the identified liner state-space model and the highly nonlinear system) is minimized. Therein, the modelling error is a function of the servo-controlled closed-loop system state and control input. Once the optimal linear model is obtained, an integrated compensation-improvement linear state observer, functional-observer-based EID estimator, and model predictive control (MPC) tracker with input constraints for the nonlinear sampled-data system with unknown matched/mismatched disturbances is newly presented in thesis. Also, an integrated compensation-improvement linear state observer, output-feedback-based EID estimator (due to its simplicity compared with the functional-observer-based EID estimator), and optimal linear quadratic tracker for the continuous-time nonlinear system with unknown matched/mismatched disturbances is newly presented. Besides, an integrated compensation-improvement nonlinear EID estimator and the newly developed model following controller for the continuous-time nonlinear system with unknown matched disturbances is also presented, which is a rival of the previous one. Illustrative examples are given to show the superiority of the proposed methods.

[1] Astrom, K. J. and Wittenmark, B.,“Computer Controlled Systems Theory and Design,” Prentice-Hall, Inc., Englewood cliffs, N. J., 1984
[2] Armstrong, B., Khatib, O., and Burdick, J., “The explicit dynamic model and inertial parameters of the PUMA 560 arm,” Proc. 1986 IEEE Conference on Robot and Automation, San Francisco, pp. 510-518, 1986.
[3] Bruder, D., David Remy, C., and Vasudevan, R., “Nonlinear system identification of soft robot dynamics using Koopman operator theory,” International Conference on Robotics and Automation (ICRA), 2019.
[4] Chong, E. K. P. and Zak, S. H., “An Introduction to Optimization,” John Wiley & Sons, Inc, 2013.
[5] David, R., Marcelo, A., Patricio, V., Javier, M., Jose, L. C. M., Victor, B., Omar, G., Patricio, R., Mayra, E., Milton, P., and Monica, H., “Control system for a robotic arm with 6 degrees of freedom for education systems,” International Conference on Automation, Robotics and Applications (ICARA), New Zealand, pp. 358-363, 2015.
[6] Ding, S., Chen, W. H., Mei, K., and Murray-Smith, D. J., “Disturbance observer design for nonlinear systems represented by input-output models,” IEEE Transactions on Industrial Electronics, vol. 67, no. 4, pp. 1222-1232, February. 2019.
[7] Ebrahimzadeh, F., Tsai, J. S. H., Liao, Y. T., Chung, M. C., Guo, S. M., Shieh, L. S., and Wang L., “A generalised optimal linear quadratic tracker with universal applications – Part 1: Continuous-time systems,” International Journal of System Science, vol. 48, pp. 376-396, 2016.
[8] Guo, S. M., Shieh, L. S., Chen, G. R., and Lin, C. F., “Effective chaotic orbit tracker: A prediction-based digital redesign approach,” IEEE Transactions on Circuits and Systems I-Fundamental Theory and Applications, vol. 47, no. 11, pp. 1557-1570, 2000.
[9] Gray, W. S., Venkatesh, G. S., and Duffaut Espinosa, L. A., “Nonlinear system identification for multivariable control via discrete-time Chen–Fliess series,” Automatica, 2020.
[10] Han, C., Ma, H., Zuo, W., Chen, S., and Zhang, X., “A general 6-DOF industrial robot arm control system based on Linux and FPGA,” Chinese Control and Decision Conference (CCDC), Shenyang, pp. 1220-1225, 2018.
[11] Juang, J. N., ‘‘Applied System Identification,” Prentice Hall, New Jersey, 1994.
[12] Kuo, Y. S., “A PID Filter-shaped PI-MPC Tracker and EID Estimator for Sampled-data Systems with Unknown Disturbances and Input Constraints,” Department of Electrical Engineering, National Cheng Kung University, Tainan, Taiwan, Thesis for Master, July 2021.
[13] Lin, B. J., “Some Insights on System Identification and Servo Control for Highly Nonlinear Systems,” Department of Electrical Engineering, National Cheng Kung University, Tainan, Taiwan, Thesis for Master, July 2019.
[14] Lewis, F. L., Abdallah, C. T., and Dawson, D. M., “Control of Robot Manipulators,” Macmillan Publishing Company, New York, 1993.
[15] Rabiee, H., Ataei, M., and Ekramian, M., “Continuous nonsingular terminal sliding mode control based on adaptive sliding mode disturbance observer for uncertain nonlinear systems,” Automatica, 2019.
[16] Tsai, J. S. H., Yu, T. H., Su, T. J., Guo, M., Shieh, L. S., and Canelon, J. I., “A novel on-line OCID method and its application to input-constrained active fault-tolerant tracker design for unknown nonlinear systems,” International Journal of Systems Science, vol. 50, no. 14, pp. 2632-2662, 2019.
[17] Tsai, J. S. H., Wang, H. H., Guo, S. M., Shieh, L. S., and Canelon, J. I., “A case study on the universal compensation-improvement mechanism: A robust PID filter-shaped optimal PI tracker for systems with/without disturbances,” Journal of the Franklin Institute, vol. 355, pp. 3583-3618, April 2018.
[18] Tsai, J. S. H., Kuo, L. Y., Guo, S. M., Shieh, L. S., and Canelon, J. I., “Application of the observer/Kalman filter identification method to unknown time-delay disturbed systems and the associated optimal digital tracker design,” International Journal of Systems Science, vol. 50, no. 16, pp. 2935-2961, November. 24, 2019.
[19] Takuya, H. and Qingjiu, H., “Control of position, attitude, force and moment of 6-DOF manipulator by impedance control,” International Conference on Control, Automation, Robotics and Vision (ICARCV), Singapore, pp. 274-279, 2018.
[20] Tang, D., Chen, L., and Hu, E., “A novel unknown-input estimator for disturbance estimation and compensation,” Proceedings of the Australasian Conference on Robotics and Automation, The University of Melbourne, Melbourne, Australia, December 2014.
[21] Tsai, J. S. H., Chang, C. Y., Chen, Y. F., Guo, S. M., Shieh, L. S., and Canelon, J. I., “A modified functional observer-based EID estimator for unknown sampled-data singular systems,” International Journal of Systems Science, vol. 50, no. 10, pp. 1976-2001, 2019.
[22] Wu, C. Y., Tsai, J. S. H., Guo, S. M., Shieh, L. S., Canelon, J. I., Ebrahimzadeh, F., and Wang, L., “A novel on-line observer/Kalman filter identification method and its application to input-constrained active fault-tolerant tracker design for unknown stochastic systems,” Journal of the Franklin Institute-Engineering and Applied Mathematics, vol. 352, pp. 1119-1151, 2015.
[23] Wu, C. Y., Tsai, J. S. H., Su, T. J., Guo, S. M., Shieh, L. S., and Yan, J. J., “Novel OCID method-based minimal realizations in block observable/controllable canonical forms and compensation improvement.” International Journal of Systems Science, vol. 48, no. 7, pp. 1522-1535, 2017.
[24] L. P. Wang, “Model Predictive Control System Design and Implementation Using MATLAB,” Springer, London, 2009.
[25] Yang, J., Li, S., and Chen, W. H., “Nonlinear disturbance observer-based control for multi-input multi-output nonlinear systems subject to mismatching condition,” International Journal of Control, vol. 85, no. 8, pp. 1071-1082, 2012.
[26] Zabinsky, Z. B., “Random Search Algorithms,”John Wiley & Sons, Inc, 2010.