Title :

DDPG based LADRC trajectory tracking control for underactuated unmanned ship under environmental disturbances

Author(s) :

Zheng, Yuemin [Auteur]
Génie des Matériaux Textiles - ULR 2461 [GEMTEX]
Tao, J. [Auteur]
Hartikainen, J. [Auteur]
Duan, F. [Auteur]
Sun, H. [Auteur]
Sun, M. W. [Auteur]
Sun, Q. L. [Auteur]
Zeng, Xianyi [Auteur]
Génie des Matériaux Textiles - ULR 2461 [GEMTEX]
Chen, Z. Q. [Auteur]
Xie, G. M. [Auteur]

Journal title :

Ocean Engineering

Abbreviated title :

Ocean Eng.

Volume number :

271

Publication date :

2023-03

ISSN :

0029-8018

English keyword(s) :

Underactuated unmanned ship
Trajectory tracking control
Guidance law
Linear active disturbance rejection control
Deep reinforcement learning
DDPG

HAL domain(s) :

Sciences de l'ingénieur [physics]

English abstract : [en]

Realizing trajectory tracking control of underactuated unmanned ships is critical for safe and stable navigation during offshore operations. The principal difficulties of this task lie in the coupling problem caused by the ...
Show more >Realizing trajectory tracking control of underactuated unmanned ships is critical for safe and stable navigation during offshore operations. The principal difficulties of this task lie in the coupling problem caused by the underactuated system and the uncertainty caused by the modeling error, internal parameter perturbations, and external environment. In this paper, a three-degree-of-freedom ship model is first built based on the manipulative modeling group equation. Then, a trajectory tracking line-of-sight guidance law is proposed to achieve the desired heading angle and forward speed during trajectory tracking, and the Lyapunov function is used to analyze its stability. Furthermore, the heading angle controller and forward speed controller are designed based on the linear active disturbance rejection control framework. In addition, to solve the challenging problem of controller parameter tuning and further improve the controller’s robustness, the Deep Deterministic Policy Gradient algorithm is applied to adjust the controller parameters. Finally, the effectiveness of the proposed method is verified by the simulation results of sinusoidal, circular, and square trajectories with wind and current disturbances.Show less >

Language :

Anglais

Audience :

Internationale

Popular science :

Non

Administrative institution(s) :

Université de Lille
ENSAIT
Junia HEI

Collections :

• Génie des matériaux textiles (GEMTEX) - ULR 2641

Submission date :

2023-06-20T12:11:35Z
2024-02-20T10:58:58Z