2023
DOI: 10.1016/j.engappai.2022.105589
|View full text |Cite
|
Sign up to set email alerts
|

A general motion controller based on deep reinforcement learning for an autonomous underwater vehicle with unknown disturbances

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
2
1

Citation Types

0
3
0

Year Published

2023
2023
2024
2024

Publication Types

Select...
7

Relationship

1
6

Authors

Journals

citations
Cited by 9 publications
(3 citation statements)
references
References 26 publications
0
3
0
Order By: Relevance
“…A torpedo-shaped underactuated UUV, which employs a cross-rudder and singlepropeller actuator layout, was chosen as the research object of this paper [23]. We define the position, attitude, and velocity of the underdriven UUV in the Earth-fixed coordinate system (O E − X E Y E Z E ) and body-axis coordinate system (O B − X B Y B Z B ), respectively.…”
Section: Underactuated Uuv Model Descriptionmentioning
confidence: 99%
“…A torpedo-shaped underactuated UUV, which employs a cross-rudder and singlepropeller actuator layout, was chosen as the research object of this paper [23]. We define the position, attitude, and velocity of the underdriven UUV in the Earth-fixed coordinate system (O E − X E Y E Z E ) and body-axis coordinate system (O B − X B Y B Z B ), respectively.…”
Section: Underactuated Uuv Model Descriptionmentioning
confidence: 99%
“…On the one hand, various improved PID combined with modern and intelligent methods is designed to improve the antidisturbance capability. On the other hand, model‐based methods such as backstepping control (Li & Lee, 2005; Tran et al, 2021), sliding mode control (SMC) (Koofigar, 2014), adaptive control (Pezeshki et al, 2016), model predictive control (Shen et al, 2017; Yao et al, 2019), neural network (Askari et al, 2022), reinforcement learning (Tong et al, 2023; Wang et al, 2024), and other Artificial Intelligence methods are proposed to enhance the adaptability and robustness. While the aforementioned literatures have demonstrated excellent depth‐tracking control performance, the majority of them heavily rely on exact mathematical models, which are typically challenging to come by in practice (Shi et al, 2017).…”
Section: Introductionmentioning
confidence: 99%
“…The use of disturbance observers has shown to be beneficial for the attenuation of disturbance effects. This is the case of the motion control strategy that integrates deep reinforcement learning and the extended state observer (ESO) to compensate for the model uncertainties into an optimal motion control policy [19]. Although the previously-mentioned controllers have effectively improve the closed-loop robustness of AUV motion control, there are opportunities to enhance the design process and results of control strategies for AUV.…”
Section: Introductionmentioning
confidence: 99%