Continual Learning of Range-Dependent Transmission Loss for Underwater Acoustic using Conditional Convolutional Neural Net

TL;DR

This work tackles the challenge of predicting far-field underwater transmission loss in oceans with range-dependent bathymetry. It introduces RC-CAN, a range-conditioned encoder–decoder that maps bathymetry to TL in a single shot, and couples it with a replay-based continual learning framework to generalize across diverse bathymetric environments. Using ground-truth TL generated by Bellhop ray/beam tracing for seven bathymetry scenarios (ideal seamounts, wedges, and Dickins Seamount), RC-CAN achieves high fidelity (SSIM ≈ 0.88–0.98) and orders-of-magnitude faster inference than traditional solvers, enabling near real-time predictive capability. The approach supports adaptive management of underwater radiated noise and serves as a data-driven digital twin for ocean acoustics, with potential extension to higher-fidelity data and on-line learning.

Abstract

There is a significant need for precise and reliable forecasting of the far-field noise emanating from shipping vessels. Conventional full-order models based on the Navier-Stokes equations are unsuitable, and sophisticated model reduction methods may be ineffective for accurately predicting far-field noise in environments with seamounts and significant variations in bathymetry. Recent advances in reduced-order models, particularly those based on convolutional and recurrent neural networks, offer a faster and more accurate alternative. These models use convolutional neural networks to reduce data dimensions effectively. However, current deep-learning models face challenges in predicting wave propagation over long periods and for remote locations, often relying on auto-regressive prediction and lacking far-field bathymetry information. This research aims to improve the accuracy of deep-learning models for predicting underwater radiated noise in far-field scenarios. We propose a novel range-conditional convolutional neural network that incorporates ocean bathymetry data into the input. By integrating this architecture into a continual learning framework, we aim to generalize the model for varying bathymetry worldwide. To demonstrate the effectiveness of our approach, we analyze our model on several test cases and a benchmark scenario involving far-field prediction over Dickin's seamount in the Northeast Pacific. Our proposed architecture effectively captures transmission loss over a range-dependent, varying bathymetry profile. This architecture can be integrated into an adaptive management system for underwater radiated noise, providing real-time end-to-end mapping between near-field ship noise sources and received noise at the marine mammal's location.

Figures (18)

• Figure 1: Illustration of real-time noise prediction from a marine vessel at a distant mammal location.
• Figure 2: A schematic of finite-difference approximation on a rectangular mesh.
• Figure 3: An illustration of the proposed range conditional convolutional network.
• Figure 4: Idealized seamount profiles. (a-c): fixed base width (20 km) and peak depth (500 m to 1600 m). (d-f): peak depth (800 m or 1200 m) and base width (5 km to 40 km). (g-i): randomly sampled peak depths and base widths.
• Figure 5: Wedge bathymetry profiles. (a) Depth decreasing profile, (b) increasing depth profile, and (c) decreasing then increasing depth profile.