Education, Science, Technology, Innovation and Life
Open Access
Sign In
  1. Home
  2. Journals
  3. Journal of Engineering Mechanics and Machinery
  4. Vol 11, Issue 1, 2026

Research on UUV Trajectory Tracking Method under the Fusion Mechanism of Enhanced Gaussian Process and MPC

Download as PDF

DOI: 10.23977/jemm.2026.110106 | Downloads: 2 | Views: 78

Author(s)

Xiaolong Fan 1, Xuyu Shen 1, Zhenzhong Chu 1, Yushen Duan 1

Affiliation(s)

1 School of Mechanical Engineering, University of Shanghai for Science and Technology, Yangpu, Shanghai, China

Corresponding Author

Zhenzhong Chu

ABSTRACT

This paper addresses the trajectory tracking control problem for Unmanned Underwater Vehicle systems under unknown dynamics and uncertainties. The goal is to develop a model predictive control (MPC) method that does not require an exact mechanistic model, overcoming the challenges posed by model mismatch in complex nonlinear systems. Traditional Gaussian process-based MPC methods, with their multi-input single-output characteristics, can only independently compensate for each model dimension, making it difficult to capture coupling relationships between multiple outputs, limiting control accuracy and robustness. To address this, a multi-output Gaussian process MPC method based on a cooperative regionalized model is proposed. This method uses a unified modeling framework to learn dynamic errors and their structure between the theoretical model and real system, effectively representing output coupling, thus improving tracking performance. Additionally, a stochastic constraint handling strategy based on Gaussian process prediction uncertainty is introduced, converting probabilistic constraints into deterministic convex constraints. This ensures efficient online solving and enhances the system's safe operation in complex environments. Finally, comparative simulations demonstrate that the proposed method outperforms classical Gaussian process MPC in terms of tracking accuracy, dynamic response, and constraint satisfaction.

KEYWORDS

Unmanned Underwater Vehicle, Model Predictive Control, Gaussian Process, Uncertainty

CITE THIS PAPER

Xiaolong Fan, Xuyu Shen, Zhenzhong Chu, Yushen Duan. Research on UUV Trajectory Tracking Method under the Fusion Mechanism of Enhanced Gaussian Process and MPC. Journal of Engineering Mechanics and Machinery (2026). Vol. 11, No. 1, 52-63. DOI: http://dx.doi.org/10.23977/jemm.2026.110106.

REFERENCES

[1] WANG N, GAO Y, ZHANG X. Data-driven performance-prescribed reinforcement learning control of an unmanned surface vehicle[J]. IEEE Transactions on Neural Networks and Learning Systems, 2021, 32(12):5456-5467.
[2] WANG N, GAO Y, ZHAO H, et al. Reinforcement learning-based optimal tracking control of an unknown unmanned surface vehicle[J]. IEEE Transactions on Neural Networks and Learning Systems, 2021, 32(7):3034-3045.
[3] PENG G, LU Z, TAN Z, et al. A novel algorithm based on nonlinear optimization for parameters calibration of wheeled robot mobile chasses[J]. Applied Mathematical Modelling, 2021, 95:396-408.
[4] YAN Z, YAN J, CAI S, et al. Robust MPC-based trajectory tracking of autonomous underwater vehicles with model uncertainty[J]. Ocean Engineering, 2023, 286:115617.
[5] DERINGER V L, BARTÓK A P, BERNSTEIN N, WILKINS D M, CERIOTTI M, CSÁNYI G. Gaussian process regression for materials and molecules[J]. Chemical Reviews, 2021, 121(16): 10073-10141.
[6] X. Yan and Y. Liu, "UUV Trajectory Tracking Control Based on Gaussian Process Model Predictive Control," 2024 3rd Conference on Fully Actuated System Theory and Applications (FASTA), Shenzhen, China, 2024. 1146-1151.
[7] SOLOPERTO R, KÖHLER J, ALLGÖWER F. Augmenting mpc schemes with active learning: Intuitive tuning and guaranteed performance[J]. IEEE Control Systems Letters, 2020, 4(3):713-718.
[8] XU X, CHEN H, LIAN C, et al. Learning-based predictive control for discrete-time nonlinear systems with stochastic disturbances[J]. IEEE Transactions on Neural Networks and Learning Systems, 2018, 29(12):6202-6213.
[9] AN Z, GONG P, ZHANG W, et al. Model predictive control of autonomous underwater vehicles for trajectory tracking with external disturbances[J]. Ocean Engineering, 2020, 217:107884.
[10] MORATO M M, FELIX M S. Data Science and Model Predictive Control: A survey of recent advances on data-driven MPC algorithms[J]. Journal of Process Control, 2024, 144: 103327.
[11] SUN D, JAMSHIDNEJAD A, DE SCHUTTER B. A novel framework combining MPC and deep reinforcement learning with application to freeway traffic control[J]. IEEE Transactions on Intelligent Transportation Systems, 2024, 25(7): 6756-6769.
[12] HEWING L, KABZAN J, ZEILINGER M N. Cautious model predictive control using gaussian process regression[J]. IEEE Transactions on Control Systems Technology, 2020, 28(6):2736-2743.
[13] A. P. Aguiar, A. M. Pascoal, Dynamic positioning and way-point tracking of underactuated AUVs in the presence of ocean currents, Proceedings of the 41st IEEE Conference on Decision and Control, (2002) 2105 2110.
[14] Qikun Shen, Peng Shi, Chee Peng Lim, Fuzzy adaptive fault-tolerant stability control against novel actuator faults and its application to me chanical systems, IEEE Transactions on Fuzzy Systems, 32 (4) (2024) 2331–2340.
[15] Wenjin Wang, Tao Wen, Xiao He, Guohua Xu, Robust trajectory tracking and control allocation of X-rudder AUV with actuator uncertainty, Control Engineering Practice, 136 (2023) 105535.
[16] L. Hewing, J. Kabzan, and M. N. Zeilinger, "Cautious model predictive control using gaussian process regression," IEEE Transactions on Control Systems Technology, vol. 28, no. 6, pp. 2736–2743, 2020.
[17] E. Schulz, M. Speekenbrink, and A. Krause, "A tutorial on gaussian process regression: Modelling, exploring, and exploiting functions," Journal of Mathematical Psychology, vol. 85, pp. 1–16, 2018.

All published work is licensed under a Creative Commons Attribution 4.0 International License.

Copyright © 2016 - 2031 Clausius Scientific Press Inc. All Rights Reserved.