本論文旨在發展一套適用於自主無人船（USV）的水動力係數辨識與強健運動控制整合系統。首先，針對傳統平面運動機構（PMM）水槽實驗的侷限性，提出以實船現地多自由度動態實驗為核心的資料蒐集流程，並利用高精度慣性導航系統（INS）於多種操控模式（直線、轉圈、zig-zag等）下取得加速度、速度、姿態等運動數據。經由濾波、特徵萃取及動態分段等資料前處理後，結合理論模型進行水動力參數辨識，並建立專屬的RBF神經網路以捕捉各項動力學係數的非線性關係。
進一步地，本文設計一套結合自適應RBF神經網路與Lyapunov穩定性分析的非線性控制器，實現系統參數在線辨識及外部擾動即時補償。控制架構理論推導完整，並嚴謹證明追蹤誤差與參數辨識誤差皆可漸進收斂，確保控制系統之強健性與穩定性。最後，透過數值模擬驗證所提模型與控制方法於不同工況下皆具備高精度軌跡追蹤與動態適應能力。
本研究成果證實，所提資料驅動之水動力參數辨識流程及自適應控制架構，可顯著提升自主無人船於複雜環境下之建模精度與運動控制效能，對智慧型海洋載具實地應用與後續技術發展具有重要參考價值。

This thesis presents an integrated framework for hydrodynamic parameter identification and robust nonlinear control of autonomous unmanned surface vessels (USVs). To overcome the limitations of traditional planar motion mechanism (PMM) tank experiments, a full-scale, field-oriented data acquisition procedure is developed, leveraging high-precision inertial navigation system (INS) measurements under various maneuvering modes (e.g., straight-line, turning circle, and zig-zag). Through systematic data preprocessing—including filtering, feature extraction, and segment selection—combined with theoretical modeling, key hydrodynamic coefficients are identified and captured using dedicated radial basis function (RBF) neural networks to characterize nonlinear dynamics.
A nonlinear adaptive controller, incorporating RBF neural network estimation and Lyapunov-based stability analysis, is then designed to achieve online parameter identification and real-time disturbance compensation. Rigorous theoretical proofs demonstrate that the proposed control scheme guarantees convergence of both tracking and identification errors, ensuring system robustness and stability. Numerical simulations confirm that the proposed modeling and control approach enables accurate trajectory tracking and dynamic adaptation under various environmental conditions.
The results demonstrate that the proposed data-driven parameter identification and adaptive control framework significantly enhances modeling accuracy and control performance of USVs in complex maritime environments, providing a valuable foundation for the practical deployment and future development of intelligent marine vehicles.

[1] 伊莎貝兒·格雷特森（Isabelle Gerretsen），「NASA美國太空總署為何不但上天還要下海？」BBC中文網，2022年2月1日。 [線上]. Available: https://www.bbc.com/zhongwen/trad/science-60125571 (accessed July 22, 2025).
[2] Allianz Global Corporate & Specialty, “It is estimated that 75% to 96% of marine accidents can involve human error.” [Online]. Available: https://www.agcs.allianz.com/news-and-insights/expert-risk-articles/human-error-shipping-safety.html (accessed June 2019).
[3] Fortune Business Insights, “The global autonomous ships market is expected to grow from USD 6.55 billion in 2021 to USD 12.07 billion in 2028 at a CAGR of 9.13%.” [Online]. Available: https://www.fortunebusinessinsights.com/industry-reports/autonomous-ship-market-101797 (accessed Feb. 2022).
[4] HD Hyundai, “HD Hyundai’s Avikus Successfully Conducts the World’s First Transoceanic Voyage of a Large Merchant Ship Relying on Autonomous Navigation Technologies,” 2022.
[5] T. I. Fossen, "Handbook of Marine Craft Hydrodynamics and Motion Control," 2011.
[6] ITTC – Specialist Committee on Energy Saving Methods, Resistance and Propulsion Test and Performance Prediction with Skin Frictional Drag Reduction Techniques, ITTC Recommended Procedures and Guidelines, 7.5-02-02-03, 28th ITTC, pp. 1–12, 2017. [Online]. Available: https://ittc.info/media/8017/75-02-02-03.pdf
[7] Martins, R. et al., "Hydrodynamic parameter identification for full-scale USVs," 2022.
[8] J. Yue, L. Liu, N. Gu, Z. Peng, D. Wang, and Y. Dong, "Online adaptive parameter identification of an unmanned surface vehicle without persistency of excitation," Ocean Engineering, vol. 250, p. 110232, Apr. 2022. doi: 10.1016/j.oceaneng.2021.110232
[9] P. F. Xu, C. Cheng, H. X. Cheng, Y. L. Shen, and Y. X. Ding, “Identification-based 3 DOF model of unmanned surface vehicle using support vector machines enhanced by cuckoo search algorithm,” Ocean Engineering, vol. 197, p. 106898, 2020. [Online]. Available: https://doi.org/10.1016/j.oceaneng.2019.106898
[10] J. Fei, Z. Wang, X. Lu, and L. Deng, "Adaptive RBF neural network control based on sliding mode controller for active power filter," in Proceedings of the 32nd Chinese Control Conference, 2013, pp. 6501–6505. doi: 10.1109/ChiCC.2013.6639988
[11] T. I. Fossen, Guidance and Control of Ocean Vehicles. Chichester, UK: Wiley, 1994.
[12] P. A. Ioannou and J. Sun, "Robust Adaptive Control," 2nd Ed., Dover Publications, 2012.
[13] C. M. Bishop, "Neural Networks for Pattern Recognition," Oxford University Press, 1995.
[14] S. Chen, C. F. N. Cowan, and P. M. Grant, "Orthogonal least squares learning algorithm for radial basis function networks," IEEE Transactions on Neural Networks, vol. 2, no. 2, pp. 302–309, 1991. [Online]. Available: https://doi.org/10.1109/72.80341
[15] S. Haykin, "Neural Networks: A Comprehensive Foundation," 2nd Ed., Prentice Hall, 1999.
[16] Chen, Y.-Y.; Tsao, C.-H.; Liu, K.-H and Huang, Y.-X.;, “CONTROL SYSTEM DESIGN FOR UNDERWATER VEHICLES WITH ACTUATOR CONSTRAINTS”. Journal of Taiwan Society of Naval Architects and Marine Engineers. vol. 41, no. 2, pp. 77-90, 2022.
[17] S. H. Żak, "Systems and Control," New York, NY, USA: Oxford University Press, 2003.
[18] Y.-Y. Chen, C.-H. Tsao, K.-H. Liu, and Y.-X. Huang, "Control system design for underwater vehicles with actuator constraints," Journal of Taiwan Society of Naval Architects and Marine Engineers, vol. 41, no. 2, pp. 77-90, 2022.
[19] A. Lovo-Ayala, R. Soto-Diaz, C. A. Gutierrez-Martinez, J. F. Jimenez-Vargas, J. Jiménez-Cabas, and J. Escorcía-Gutierrez, “Simplified Model Characterization and Control of an Unmanned Surface Vehicle,” Journal of Marine Science and Engineering, vol. 13, no. 4, p. 813, Apr. 2025. [Online]. Available: https://doi.org/10.3390/jmse13040813
[20] Z. Han, Y. Wang, and Q. Sun, “Straight-Path Following and Formation Control of USVs Using Distributed Deep Reinforcement Learning and Adaptive Neural Network,” IEEE/CAA Journal of Automatica Sinica, vol. 10, no. 2, pp. 572–574, Feb. 2023. [Online]. Available: https://doi.org/10.1109/JAS.2023.123255
[21] H. Qin, X. Chen, and Y. Sun, "Adaptive state-constrained trajectory tracking control of unmanned surface vessel with actuator saturation based on RBFNN and tan-type barrier Lyapunov function," Ocean Engineering, vol. 253, 2022, Art. no. 110966. [Online]. Available: https://doi.org/10.1016/j.oceaneng.2022.110966
[22] J. Amini Foroushani and M. Sabzpooshani, "An approach for the estimation of hydrodynamic coefficients of an underwater vehicle in off-design velocities based on an improved RBF neural network," Journal of Marine Science and Technology, vol. 25, pp. 1–18, 2020. [Online]. Available: https://doi.org/10.1007/s00773-020-00740-7
[23] D. Liu, J. Liu, C. Sun, and B. Dai, "Convex Optimization-Based Adaptive Neural Network Control for Unmanned Surface Vehicles Considering Moving Obstacles," Journal of Marine Science and Engineering, vol. 13, no. 3, p. 587, 2025. [Online]. Available: https://doi.org/10.3390/jmse13030587
[24] Y. Chen and H. Chen, "Prescribed performance control of underactuated surface vessels’ trajectory using a neural network and integral time-delay sliding mode," Kybernetika, vol. 59, no. 2, pp. 273–293, 2023. [Online]. Available: https://doi.org/10.14736/kyb-2023-2-0273
[25] H. Bao, H. Zhu, and D. Liu, "Research on hydrodynamic modeling and simulation of streamlined autonomous underwater vehicle based on CFD method," in Proc. 2023 Chinese Control and Decision Conference (CCDC), 2023. [Online]. Available: https://doi.org/10.1109/CCDC58219.2023.10327061
[26] J. Li, G. Zhang, and B. Li, "Robust adaptive neural cooperative control for the USV-UAV based on the LVS-LVA guidance principle," Journal of Marine Science and Engineering, vol. 10, no. 1, p. 51, 2022. [Online]. Available: https://doi.org/10.3390/jmse10010051
[27] X. Wang, X. Yin, and F. Shen, "Robust adaptive neural tracking control for a class of nonlinear systems with unmodeled dynamics using disturbance observer," Neurocomputing, vol. 292, pp. 49–62, 2018. [Online]. Available: https://doi.org/10.1016/j.neucom.2018.02.082