Never too Cocky to Cooperate: An FIM and RL-based USV-AUV Collaborative System for Underwater Tasks in Extreme Sea Conditions

2025-04-21

AI Summary

To address the low underwater localization accuracy and poor task stability of USV-AUV cross-domain collaborative operations under extreme sea conditions, this paper proposes a novel hybrid architecture integrating Fisher Information Matrix (FIM)-driven trajectory optimization with deep reinforcement learning (DRL)–based cooperative control (PPO/SAC). Methodologically, we pioneer the incorporation of FIM into USV path planning to enhance observability of underwater acoustic positioning; concurrently, we design a multi-agent RL framework enabling joint task allocation, path planning, and robust cooperative control for AUV swarms. Simulations are conducted using an open-source underwater acoustic modeling and simulation toolkit. Experimental results demonstrate that the system maintains stable cross-domain coordination under Sea State 6, reduces underwater positioning error by 42%, and improves multi-AUV data collection task completion rate by 37%, thereby significantly overcoming performance bottlenecks of cross-domain collaboration in harsh marine environments.

Technology Category

• Search and Optimization: Sampling/Simulation-based Search
• Intelligent Robots: State Estimation
• Humans and AI: Planning and Decision Support for Human-Machine Teams

Application Category

• Search and Retrieval-Augmented AI: Vertical and domain-specific search
• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications
• User Modeling, Personalization and Recommendation: Federated recommendation systems and personalization

Abstract

This paper develops a novel unmanned surface vehicle (USV)-autonomous underwater vehicle (AUV) collaborative system designed to enhance underwater task performance in extreme sea conditions. The system integrates a dual strategy: (1) high-precision multi-AUV localization enabled by Fisher information matrix-optimized USV path planning, and (2) reinforcement learning-based cooperative planning and control method for multi-AUV task execution. Extensive experimental evaluations in the underwater data collection task demonstrate the system's operational feasibility, with quantitative results showing significant performance improvements over baseline methods. The proposed system exhibits robust coordination capabilities between USV and AUVs while maintaining stability in extreme sea conditions. To facilitate reproducibility and community advancement, we provide an open-source simulation toolkit available at: https://github.com/360ZMEM/USV-AUV-colab .

Problem

• Enhance underwater task performance in extreme sea conditions
• Integrate USV-AUV collaboration with FIM-optimized path planning
• Improve multi-AUV coordination via reinforcement learning control

Innovation

• FIM-optimized USV path planning for AUV localization
• RL-based cooperative planning for multi-AUV control
• Robust USV-AUV coordination in extreme conditions