High-Precision Surfacing Position Prediction for Underwater Gliders via Coordinate Transformation

Journal of Marine Science and Engineering, Год журнала: 2025, Номер 13(4), С. 760 - 760

Опубликована: Апрель 11, 2025

The accurate prediction of the surfacing position underwater gliders (UGs) is critical for mission success and cost-effective retrieval. However, current state-of-the-art (SOTA) methods often rely on complex multi-model integrations or large volumes ocean data, thereby increasing operational costs system complexity. In this study, we systematically introduce—for first time—a coordinate-transformation-based framework, originally applied in other navigation contexts, into UG surfacing-position-prediction task. By projecting both glider’s entry positions a Universal Transverse Mercator (UTM) planar coordinate treating resulting displacement as target, avoid dependence heavily parameterized models, simplify training process, maintain robust predictive accuracy. Our approach combines common machine learning predictors (e.g., AdaBoost, LGBM, gradient boosting, random forest, decision trees) instead advanced deep architectures, thus reducing computational overhead. Experiments two real-world sea trial datasets (containing 2159 1456 profiles, respectively) show that, compared with direct regression approaches, method improves positioning accuracy by up to 50% within 500-meter range, yet requires minimal multi-source data. Overall, study integrates concept transformation task predicting gliders, effectively streamlining without sacrificing result highly flexible approach, providing theoretical support future optimizations glider systems.

Язык: Английский

Modelling and analysis of depth-averaged currents for underwater gliders

Ocean Engineering, Год журнала: 2024, Номер 312, С. 119086 - 119086

Опубликована: Сен. 6, 2024

Язык: Английский