Heterogeneous Multi-Agent Task Planning Method in Complex Marine Environment

To enable collaborative scouting / strike / assessment of underwater time-sensitive targets by heterogeneous multi-agent systems, in this study a heterogeneous multi-agent collaborative decision-making method is proposed based on deep reinforcement learning. The method integrates two core learning f...

Full description

Saved in:
Bibliographic Details
Main Authors: Shoumin Wang, Ning Niu, Zhichao Wang, Yaxuan Lv, Jing Zhang
Format: Article
Language:English
Published: IEEE 2025-01-01
Series:IEEE Access
Subjects:
Online Access:https://ieeexplore.ieee.org/document/10988794/
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1850153186421112832
author Shoumin Wang
Ning Niu
Zhichao Wang
Yaxuan Lv
Jing Zhang
author_facet Shoumin Wang
Ning Niu
Zhichao Wang
Yaxuan Lv
Jing Zhang
author_sort Shoumin Wang
collection DOAJ
description To enable collaborative scouting / strike / assessment of underwater time-sensitive targets by heterogeneous multi-agent systems, in this study a heterogeneous multi-agent collaborative decision-making method is proposed based on deep reinforcement learning. The method integrates two core learning frameworks one for heterogeneous multi-agent task allocation and one for single-agent multi-task learning. The task allocation framework combines the proximal policy optimization algorithm with experience replay to train an Actor network, ensuring stable iterative updates of task allocation policies toward high-reward directions. Concurrently, a Critic network is trained using successor features and temporal difference methods, establishing a foundation for transfer learning. A region partitioning mechanism is introduced to construct a foundational knowledge base, enabling the transfer of sub-region knowledge acquired by multi-agents to target regions, thereby addressing complex underwater task allocation scenarios. The single-agent multi-task learning framework employs an experience replay pool with knowledge transfer attributes and policy distillation technology, allowing each agent to assimilate task-specific expertise from heterogeneous peers across diverse mission scenarios. This capability supports multi-task execution, including path planning, emergency obstacle avoidance, and trajectory tracking.
format Article
id doaj-art-be1e01601c7b474ea0cdd5ad2aa71be9
institution OA Journals
issn 2169-3536
language English
publishDate 2025-01-01
publisher IEEE
record_format Article
series IEEE Access
spelling doaj-art-be1e01601c7b474ea0cdd5ad2aa71be92025-08-20T02:25:47ZengIEEEIEEE Access2169-35362025-01-0113842028421610.1109/ACCESS.2025.356750310988794Heterogeneous Multi-Agent Task Planning Method in Complex Marine EnvironmentShoumin Wang0Ning Niu1https://orcid.org/0000-0002-6431-2279Zhichao Wang2https://orcid.org/0009-0007-5244-2301Yaxuan Lv3Jing Zhang4https://orcid.org/0000-0001-6361-6898School of Mechatronic Engineering, Xi’an Technological University, Xi’an, ChinaSchool of Electronic Science and Technology, Hainan University, Haikou, ChinaThe 32033 Troops of China People’s Liberation Army, Haikou, ChinaSchool of Electronic Science and Technology, Hainan University, Haikou, ChinaSchool of Electronic Science and Technology, Hainan University, Haikou, ChinaTo enable collaborative scouting / strike / assessment of underwater time-sensitive targets by heterogeneous multi-agent systems, in this study a heterogeneous multi-agent collaborative decision-making method is proposed based on deep reinforcement learning. The method integrates two core learning frameworks one for heterogeneous multi-agent task allocation and one for single-agent multi-task learning. The task allocation framework combines the proximal policy optimization algorithm with experience replay to train an Actor network, ensuring stable iterative updates of task allocation policies toward high-reward directions. Concurrently, a Critic network is trained using successor features and temporal difference methods, establishing a foundation for transfer learning. A region partitioning mechanism is introduced to construct a foundational knowledge base, enabling the transfer of sub-region knowledge acquired by multi-agents to target regions, thereby addressing complex underwater task allocation scenarios. The single-agent multi-task learning framework employs an experience replay pool with knowledge transfer attributes and policy distillation technology, allowing each agent to assimilate task-specific expertise from heterogeneous peers across diverse mission scenarios. This capability supports multi-task execution, including path planning, emergency obstacle avoidance, and trajectory tracking.https://ieeexplore.ieee.org/document/10988794/Autonomous underwater vehiclecollaboration decision-makingexperiences replay methodproximal policy optimization algorithmsuccessor featureuncrewed surface vessels (USVs)
spellingShingle Shoumin Wang
Ning Niu
Zhichao Wang
Yaxuan Lv
Jing Zhang
Heterogeneous Multi-Agent Task Planning Method in Complex Marine Environment
IEEE Access
Autonomous underwater vehicle
collaboration decision-making
experiences replay method
proximal policy optimization algorithm
successor feature
uncrewed surface vessels (USVs)
title Heterogeneous Multi-Agent Task Planning Method in Complex Marine Environment
title_full Heterogeneous Multi-Agent Task Planning Method in Complex Marine Environment
title_fullStr Heterogeneous Multi-Agent Task Planning Method in Complex Marine Environment
title_full_unstemmed Heterogeneous Multi-Agent Task Planning Method in Complex Marine Environment
title_short Heterogeneous Multi-Agent Task Planning Method in Complex Marine Environment
title_sort heterogeneous multi agent task planning method in complex marine environment
topic Autonomous underwater vehicle
collaboration decision-making
experiences replay method
proximal policy optimization algorithm
successor feature
uncrewed surface vessels (USVs)
url https://ieeexplore.ieee.org/document/10988794/
work_keys_str_mv AT shouminwang heterogeneousmultiagenttaskplanningmethodincomplexmarineenvironment
AT ningniu heterogeneousmultiagenttaskplanningmethodincomplexmarineenvironment
AT zhichaowang heterogeneousmultiagenttaskplanningmethodincomplexmarineenvironment
AT yaxuanlv heterogeneousmultiagenttaskplanningmethodincomplexmarineenvironment
AT jingzhang heterogeneousmultiagenttaskplanningmethodincomplexmarineenvironment