Scalable Aerial GNSS Localization for Marine Robots (2505.04095v1)

Abstract: Accurate localization is crucial for water robotics, yet traditional onboard Global Navigation Satellite System (GNSS) approaches are difficult or ineffective due to signal reflection on the water's surface and its high cost of aquatic GNSS receivers. Existing approaches, such as inertial navigation, Doppler Velocity Loggers (DVL), SLAM, and acoustic-based methods, face challenges like error accumulation and high computational complexity. Therefore, a more efficient and scalable solution remains necessary. This paper proposes an alternative approach that leverages an aerial drone equipped with GNSS localization to track and localize a marine robot once it is near the surface of the water. Our results show that this novel adaptation enables accurate single and multi-robot marine robot localization.

Scalable Aerial GNSS Localization for Marine Robots

The paper "Scalable Aerial GNSS Localization for Marine Robots" presents an innovative approach to address the key challenge of efficient localization for marine robots operating near the surface of the water. Traditional methods of localization for marine robotics, such as GNSS receivers, are often hindered by reflections on water surfaces and the high costs associated with aquatic GNSS systems. Furthermore, alternatives like inertial navigation systems and acoustic methods typically suffer from complications such as error accumulation and computational complexity. This paper proposes a novel method leveraging aerial drones with GNSS localization to track and localize marine robots from above the surface, providing a scalable solution adaptable to both single and multi-robot scenarios.

Proposed Methodology

Central to the paper is the integration of drone-based GNSS systems, which have become affordable and widely used. The authors propose using such drones to visually track marine robots, leveraging their position and movement data captured by vision models. The proposed algorithm for localization involves three key components:

1. Data Acquisition: The drone captures images of the marine robot when it is close to the surface, along with metadata like camera angles, drone direction, and image parameters. These serve as the basis for the visual localization process.
2. Visual Localization: Employing a convolution-based vision model, specifically YOLOv11, the algorithm focuses on object detection and tracking. The model is optimized through data augmentation techniques to handle challenging oceanic environments.
3. Estimation: Utilizing the data from visual localization, the algorithm computes GNSS coordinates for the marine robots, considering the drone's altitude and camera specifications.

Experimental Validation

Experiments demonstrate that the proposed method achieves localization accuracy with errors varying from 0.5m to 4.5m under different conditions. The tests indicate that localization errors generally diminish when the drone is positioned closer to the marine robots. Although sensor noise, detection uncertainty, tides, and currents introduce minor discrepancies, these errors are within expected GNSS accuracy ranges, substantiating the practicality of the aerial GNSS localization approach.

Implications and Future Directions

This work has significant implications for underwater exploration, environmental monitoring, and marine operations where accurate localization is pivotal. By avoiding the need for expensive, specialized GNSS systems for each marine robot, the drone-based method facilitates cost-effective and scalable solutions for fleet operations.

Potential areas for future research include refining the accuracy of the model by incorporating additional drones to cross-validate positional estimations or optimizing sensor calibration for enhanced detection precision. Collaborative position estimation using multiple drones represents another promising direction, possibly enabling even greater localization accuracy and operational efficiency in complex marine environments.

In summary, the paper provides a compelling case for aerial GNSS localization in marine robotics, paving the way for advancements in drone-assisted marine navigation strategies and technology-driven ocean exploration initiatives.