MarineFormer: A Spatio-Temporal Attention Model for USV Navigation in Dynamic Marine Environments

TL;DR

MarineFormer is a novel Transformer-based model that effectively fuses flow data with sensor inputs to enhance autonomous USV navigation in complex, dynamic marine environments, significantly improving success rates and path efficiency.

Contribution

This work introduces MarineFormer, a new attention-based architecture that integrates flow measurements with sensor data for improved navigation in challenging marine scenarios.

Findings

Improves success rate by nearly 23% over baselines.

Reduces path length in navigation tasks.

Flow data integration is critical for performance.

Abstract

Autonomous navigation in marine environments can be extremely challenging, especially in the presence of spatially varying flow disturbances and dynamic and static obstacles. In this work, we demonstrate that incorporating local flow field measurements fundamentally alters the nature of the problem, transforming otherwise unsolvable navigation scenarios into tractable ones. However, the mere availability of flow data is not sufficient; it must be effectively fused with conventional sensory inputs such as ego-state and obstacle states. To this end, we propose \textbf{MarineFormer}, a Transformer-based policy architecture that integrates two complementary attention mechanisms: spatial attention for sensor fusion, and temporal attention for capturing environmental dynamics. MarineFormer is trained end-to-end via reinforcement learning in a 2D simulated environment with realistic flow features and obstacles. Extensive evaluations against classical and state-of-the-art baselines show that our approach improves episode completion success rate by nearly 23\% while reducing path length. Ablation studies further highlight the critical role of flow measurements and the effectiveness of our proposed architecture in leveraging them.