Saved in:
Bibliographic Details
Main Authors: Feng, Yukai, Wu, Zhiheng, Wu, Zhengxing, Gu, Junwen, Yu, Junzhi
Format: Preprint
Published: 2026
Subjects:
Online Access:https://arxiv.org/abs/2604.19404
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866915947423465472
author Feng, Yukai
Wu, Zhiheng
Wu, Zhengxing
Gu, Junwen
Yu, Junzhi
author_facet Feng, Yukai
Wu, Zhiheng
Wu, Zhengxing
Gu, Junwen
Yu, Junzhi
contents Traditional policy learning methods in cooperative pursuit face fundamental challenges in biomimetic underwater robots, where long-horizon decision making, partial observability, and inter-robot coordination require both expressiveness and stability. To address these issues, a novel framework called Mamba-based multi-agent group relative policy optimization (M$^{2}$GRPO) is proposed, which integrates a selective state-space Mamba policy with group-relative policy optimization under the centralized-training and decentralized-execution (CTDE) paradigm. Specifically, the Mamba-based policy leverages observation history to capture long-horizon temporal dependencies and exploits attention-based relational features to encode inter-agent interactions, producing bounded continuous actions through normalized Gaussian sampling. To further improve credit assignment without sacrificing stability, the group-relative advantages are obtained by normalizing rewards across agents within each episode and optimized through a multi-agent extension of GRPO, significantly reducing the demand for training resources while enabling stable and scalable policy updates. Extensive simulations and real-world pool experiments across team scales and evader strategies demonstrate that M$^{2}$GRPO consistently outperforms MAPPO and recurrent baselines in both pursuit success rate and capture efficiency. Overall, the proposed framework provides a practical and scalable solution for cooperative underwater pursuit with biomimetic robot systems.
format Preprint
id arxiv_https___arxiv_org_abs_2604_19404
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle M$^{2}$GRPO: Mamba-based Multi-Agent Group Relative Policy Optimization for Biomimetic Underwater Robots Pursuit
Feng, Yukai
Wu, Zhiheng
Wu, Zhengxing
Gu, Junwen
Yu, Junzhi
Robotics
Artificial Intelligence
Traditional policy learning methods in cooperative pursuit face fundamental challenges in biomimetic underwater robots, where long-horizon decision making, partial observability, and inter-robot coordination require both expressiveness and stability. To address these issues, a novel framework called Mamba-based multi-agent group relative policy optimization (M$^{2}$GRPO) is proposed, which integrates a selective state-space Mamba policy with group-relative policy optimization under the centralized-training and decentralized-execution (CTDE) paradigm. Specifically, the Mamba-based policy leverages observation history to capture long-horizon temporal dependencies and exploits attention-based relational features to encode inter-agent interactions, producing bounded continuous actions through normalized Gaussian sampling. To further improve credit assignment without sacrificing stability, the group-relative advantages are obtained by normalizing rewards across agents within each episode and optimized through a multi-agent extension of GRPO, significantly reducing the demand for training resources while enabling stable and scalable policy updates. Extensive simulations and real-world pool experiments across team scales and evader strategies demonstrate that M$^{2}$GRPO consistently outperforms MAPPO and recurrent baselines in both pursuit success rate and capture efficiency. Overall, the proposed framework provides a practical and scalable solution for cooperative underwater pursuit with biomimetic robot systems.
title M$^{2}$GRPO: Mamba-based Multi-Agent Group Relative Policy Optimization for Biomimetic Underwater Robots Pursuit
topic Robotics
Artificial Intelligence
url https://arxiv.org/abs/2604.19404