An Applied Study of Improved Deep Q-Networks for Marine Path Planning

Pengcheng Li; Bingyang Liang; Qiang Ren; Yuanguo Zhou

doi:10.1109/CSRSWTC60855.2023.10427271

An Applied Study of Improved Deep Q-Networks for Marine Path Planning

Pengcheng Li
, Bingyang Liang
, Qiang Ren
, Yuanguo Zhou^*

^*Corresponding author for this work

School of Electronic and Information Engineering

Xi'an University of Science and Technology

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

Abstract

This study tackles challenges in maritime path planning, a key concern in industries like shipping and offshore oil, by introducing an enhanced Deep Q-Network (DQN) that incorporates Navigational Priority (NP) and Priority Experience Replay (PER). Utilizing a Gym-based simulation, the enhanced DQN outperforms traditional and baseline methods in both path optimization and computational efficiency, validating the use of Priority Experience Replay and suggesting new avenues for addressing complex problems.

Original language	English
Title of host publication	Proceedings - 2023 Cross Strait Radio Science and Wireless Technology Conference, CSRSWTC 2023
Publisher	Institute of Electrical and Electronics Engineers Inc.
ISBN (Electronic)	9798350358971
DOIs	https://doi.org/10.1109/CSRSWTC60855.2023.10427271
State	Published - 2023
Event	2023 Cross Strait Radio Science and Wireless Technology Conference, CSRSWTC 2023 - Guilin, China Duration: 10 Nov 2023 → 13 Nov 2023

Publication series

Name	Proceedings - 2023 Cross Strait Radio Science and Wireless Technology Conference, CSRSWTC 2023

Conference

Conference	2023 Cross Strait Radio Science and Wireless Technology Conference, CSRSWTC 2023
Country/Territory	China
City	Guilin
Period	10/11/23 → 13/11/23

UN SDGs

This output contributes to the following UN Sustainable Development Goals (SDGs)

SDG 14 Life Below Water

Keywords

Improved deep Q-network
Maritime simulation environment
Navigational Priority
Obstacle avoidance mechanism

Access to Document

10.1109/CSRSWTC60855.2023.10427271

Cite this

Li, P., Liang, B., Ren, Q., & Zhou, Y. (2023). An Applied Study of Improved Deep Q-Networks for Marine Path Planning. In Proceedings - 2023 Cross Strait Radio Science and Wireless Technology Conference, CSRSWTC 2023 (Proceedings - 2023 Cross Strait Radio Science and Wireless Technology Conference, CSRSWTC 2023). Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/CSRSWTC60855.2023.10427271

Li, Pengcheng ; Liang, Bingyang ; Ren, Qiang et al. / An Applied Study of Improved Deep Q-Networks for Marine Path Planning. Proceedings - 2023 Cross Strait Radio Science and Wireless Technology Conference, CSRSWTC 2023. Institute of Electrical and Electronics Engineers Inc., 2023. (Proceedings - 2023 Cross Strait Radio Science and Wireless Technology Conference, CSRSWTC 2023).

@inproceedings{dc3df62583084fefa79fc1f79343db85,

title = "An Applied Study of Improved Deep Q-Networks for Marine Path Planning",

abstract = "This study tackles challenges in maritime path planning, a key concern in industries like shipping and offshore oil, by introducing an enhanced Deep Q-Network (DQN) that incorporates Navigational Priority (NP) and Priority Experience Replay (PER). Utilizing a Gym-based simulation, the enhanced DQN outperforms traditional and baseline methods in both path optimization and computational efficiency, validating the use of Priority Experience Replay and suggesting new avenues for addressing complex problems.",

keywords = "Improved deep Q-network, Maritime simulation environment, Navigational Priority, Obstacle avoidance mechanism",

author = "Pengcheng Li and Bingyang Liang and Qiang Ren and Yuanguo Zhou",

note = "Publisher Copyright: {\textcopyright} 2023 IEEE.; 2023 Cross Strait Radio Science and Wireless Technology Conference, CSRSWTC 2023 ; Conference date: 10-11-2023 Through 13-11-2023",

year = "2023",

doi = "10.1109/CSRSWTC60855.2023.10427271",

language = "英语",

series = "Proceedings - 2023 Cross Strait Radio Science and Wireless Technology Conference, CSRSWTC 2023",

publisher = "Institute of Electrical and Electronics Engineers Inc.",

booktitle = "Proceedings - 2023 Cross Strait Radio Science and Wireless Technology Conference, CSRSWTC 2023",

address = "美国",

}

Li, P, Liang, B, Ren, Q & Zhou, Y 2023, An Applied Study of Improved Deep Q-Networks for Marine Path Planning. in Proceedings - 2023 Cross Strait Radio Science and Wireless Technology Conference, CSRSWTC 2023. Proceedings - 2023 Cross Strait Radio Science and Wireless Technology Conference, CSRSWTC 2023, Institute of Electrical and Electronics Engineers Inc., 2023 Cross Strait Radio Science and Wireless Technology Conference, CSRSWTC 2023, Guilin, China, 10/11/23. https://doi.org/10.1109/CSRSWTC60855.2023.10427271

An Applied Study of Improved Deep Q-Networks for Marine Path Planning. / Li, Pengcheng; Liang, Bingyang; Ren, Qiang et al.
Proceedings - 2023 Cross Strait Radio Science and Wireless Technology Conference, CSRSWTC 2023. Institute of Electrical and Electronics Engineers Inc., 2023. (Proceedings - 2023 Cross Strait Radio Science and Wireless Technology Conference, CSRSWTC 2023).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

TY - GEN

T1 - An Applied Study of Improved Deep Q-Networks for Marine Path Planning

AU - Li, Pengcheng

AU - Liang, Bingyang

AU - Ren, Qiang

AU - Zhou, Yuanguo

PY - 2023

Y1 - 2023

N2 - This study tackles challenges in maritime path planning, a key concern in industries like shipping and offshore oil, by introducing an enhanced Deep Q-Network (DQN) that incorporates Navigational Priority (NP) and Priority Experience Replay (PER). Utilizing a Gym-based simulation, the enhanced DQN outperforms traditional and baseline methods in both path optimization and computational efficiency, validating the use of Priority Experience Replay and suggesting new avenues for addressing complex problems.

AB - This study tackles challenges in maritime path planning, a key concern in industries like shipping and offshore oil, by introducing an enhanced Deep Q-Network (DQN) that incorporates Navigational Priority (NP) and Priority Experience Replay (PER). Utilizing a Gym-based simulation, the enhanced DQN outperforms traditional and baseline methods in both path optimization and computational efficiency, validating the use of Priority Experience Replay and suggesting new avenues for addressing complex problems.

KW - Improved deep Q-network

KW - Maritime simulation environment

KW - Navigational Priority

KW - Obstacle avoidance mechanism

UR - https://www.scopus.com/pages/publications/85186511729

U2 - 10.1109/CSRSWTC60855.2023.10427271

DO - 10.1109/CSRSWTC60855.2023.10427271

M3 - 会议稿件

AN - SCOPUS:85186511729

T3 - Proceedings - 2023 Cross Strait Radio Science and Wireless Technology Conference, CSRSWTC 2023

BT - Proceedings - 2023 Cross Strait Radio Science and Wireless Technology Conference, CSRSWTC 2023

PB - Institute of Electrical and Electronics Engineers Inc.

T2 - 2023 Cross Strait Radio Science and Wireless Technology Conference, CSRSWTC 2023

Y2 - 10 November 2023 through 13 November 2023

ER -

Li P, Liang B, Ren Q, Zhou Y. An Applied Study of Improved Deep Q-Networks for Marine Path Planning. In Proceedings - 2023 Cross Strait Radio Science and Wireless Technology Conference, CSRSWTC 2023. Institute of Electrical and Electronics Engineers Inc. 2023. (Proceedings - 2023 Cross Strait Radio Science and Wireless Technology Conference, CSRSWTC 2023). doi: 10.1109/CSRSWTC60855.2023.10427271

An Applied Study of Improved Deep Q-Networks for Marine Path Planning

Abstract

Publication series

Conference

UN SDGs

Keywords

Access to Document

Other files and links

Fingerprint

Cite this