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!
Description
Summary: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.
ISSN:2169-3536