本研究主要為發展輪式倒單擺機器人之系統鑑別，並藉由鑑別出的系統設計準確的控制器，進而提升系統的控制性能。由於本研究所使用的輪式倒單擺機器人的具有不穩定的平衡點，故本研究採用一種本實驗室改良的閉迴路鑑別的方法，有別於開迴路系統鑑別，可以在使用控制器讓系統保持平衡的同時鑑別系統動態，而且此方法還可以直接鑑別出參數化的系統模型。由於此方法只是於單輸入單輸出的系統，故本研究將在平衡點線性化後的系統簡化成內部及外部關於俯仰角及輪速的控制迴路分別鑑別，並別分設計控制器。透過上述的方法，本研究確實鑑別出理論系統輸出和真實輸出更接近的系統，並驗證了鑑別出的系統有滿足使參數正確收斂的條件，並在重新設計控制後提升控制性能。

This study focuses on the development of system identification for a wheeled inverted pendulum robot and aims to enhance the control performance by designing an accurate controller based on the identified system. As the wheeled inverted pendulum robot used in this study has an unstable equilibrium point, a modified closed-loop identification method developed in our laboratory is employed. Unlike open-loop system identification, this method allows for the identification of system dynamics while using a controller to maintain system balance.Moreover, this approach can directly identify parameterized system models. Since this method is applicable only to single-input-single-output systems, the study splits the system into internal and external loops concerning pitch angle and wheel speed after linearization around the equilibrium point. The identification and controller design are performed separately for these loops. Through these methods, the study successfully identifies a system that closely matches the theoretical system output with the real output. The identified system is verified to meet the conditions for accurate parameter convergence. After the redesign of the controller, there is an improvement in control performance.

摘要............i
Abstract.......ii
誌謝...........iii
目錄............iv
圖目錄..........vi
1.緒論...........1
1.1研究動機與目的.1
1.2文獻回顧.......2
1.3論文簡介.......3
2.系統分析........5
2.1系統定義及符號..5
2.2系統動態分析....7
2.3簡化系統........9
3.系統鑑別........12
3.1系統鑑別架構....12
3.2參數估測演算法..13
3.3偽隨機二進位數列.15
3.4實驗架構........16
4.硬體介紹.........18
4.1硬體設備........18
4.2通訊架構........20
5.實驗結果.........22
5.1內迴路鑑別實驗...22
5.2外迴路鑑別實驗...24
6.感測融合..........27
6.1姿態量測分析......27
6.2估測器...........29
6.3模擬環境架設......31
6.4雙IMU感測融合模擬結果.34
7.結論與未來工作.....35
7.1結論.............35
7.2未來工作.........36
8.附錄..............38
8.1系統參數.........38
8.2簡化系統參數......39
參考文獻.............40

[1] F. Grasser, A. D'Arrigo, S. Colombi and A. C. Rufer, "JOE: a mobile, inverted pendulum," in IEEE Transactions on Industrial Electronics, vol. 49, no. 1, pp. 107-114, Feb. 2002, doi: 10.1109/41.982254.
[2] Y.-S. Ha, "Trajectory tracking control for navigation of the inverse pendulum type self-contained mobile robot," Robotics and autonomous systems, vol. 17, no. 1-2, pp. 65-80, 1996.
[3] A. Salerno and J. Angeles, "The control of semi-autonomous two-wheeled robots undergoing large payload-variations," IEEE International Conference on Robotics and Automation, 2004. Proceedings. ICRA '04. 2004, New Orleans, LA, USA, 2004, pp. 1740-1745 Vol.2, doi: 10.1109/ROBOT.2004.1308075.
[4] S. W. Nawawi, M. N. Ahmad and J. H. S. Osman, "Development of a Two-Wheeled Inverted Pendulum Mobile Robot," 2007 5th Student Conference on Research and Development, Selangor, Malaysia, 2007, pp. 1-5, doi: 10.1109/SCORED.2007.4451379.
[5] N. Minouchehr and S. K. Hosseini-Sani, "Design of Model Predictive Control of two-wheeled inverted pendulum robot," 2015 3rd RSI International Conference on Robotics and Mechatronics (ICROM), Tehran, Iran, 2015, pp. 456-462, doi: 10.1109/ICRoM.2015.7367827.
[6] F. Oliveira e Silva and L. H. de Carvalho Ferreira, "Design and implementation of a PID control system for a coaxial two-wheeled mobile robot," 2013 IEEE International Symposium on Industrial Electronics, Taipei, Taiwan, 2013, pp. 1-6, doi: 10.1109/ISIE.2013.6563736.
[7] C. Yang, Z. Li, R. Cui and B. Xu, "Neural Network-Based Motion Control of an Underactuated Wheeled Inverted Pendulum Model," in IEEE Transactions on Neural Networks and Learning Systems, vol. 25, no. 11, pp. 2004-2016, Nov. 2014, doi: 10.1109/TNNLS.2014.2302475.
[8] 徐世金, "人型輪式倒單擺機器人之防打滑移動平衡控制," 清華大學動力機械工程學系學位論文, pp. 1-74, 2019.
[9] 陳藝夫, "輪式倒單擺人型機器人之訊號融合與姿態控制", 清華大學動力機械工程學系學位論文, pp. 1-59, 2020.
[10] 黃建智. "基於參數化模型之閉迴路系統鑑別 應用於多軸磁浮系統." (2020): 1-74.
[11] K. Pathak, J. Franch, and S. K. Agrawal, "Velocity and position control of a wheeled inverted pendulum by partial feedback linearization," IEEE Transactions on robotics, vol. 21, no. 3, pp. 505-513, 2005.
[12] H. Goldstein, C. Poole, and J. Safko, "Classical mechanics," ed: American Association of Physics Teachers, 2002.
[13] M. Flannery, "The enigma of nonholonomic constraints," American Journal of Physics, vol. 73, no. 3, pp. 265-272, 2005.
[14] Landau, I. D., & Boumaiza, K. (1996). An output error recursive algorithm for unbiased identification in closed loop. IFAC Proceedings Volumes, 29(1), 4144-4149.
[15] STMicroelectronics, NUCLEO F446RE, [Online]. Available: https://www.st.com/en/evaluation-tools/nucleo-f446re.html
[16] Movella, MTi-2, [Online]. Available: https://www.xsens.com/mti-2
[17] ams-OSRAM, AS5311, [Online]. Available: https://ams.com/en/as5311
[18] ROBOTIS, MX-106T, [Online]. Available: https://emanual.robotis.com/docs/en/dxl/mx/mx-106-2/
[19] C.-F. Huang, B.-H. Dai, and T.-J. Yeh, "Observer-based sensor fusion for power-assist electric bicycles," in 2016 American Control Conference (ACC), 2016, pp. 5486-5491: IEEE.
[20] Boston Dynamics, Handle, [Online]. Available: https://www.youtube.com/watch?v=-7xvqQeoA8c
[21] C.-F. Huang, B.-H. Dai, and T.-J. Yeh, "Observer-based sensor fusion for power-assist electric bicycles," in 2016 American Control Conference (ACC), 2016, pp. 5486-5491: IEEE.