Physics‐Based Learning Models for Ship Hydrodynamics

Abstract

We present the concepts of physics-based learning models (PBLM) and their relevance and application to the field of ship hydrodynamics. The utility of physics-based learning is motivated by contrasting generic learning models (GLM) for regression predictions which do not presume any knowledge of the system other than the training data provided, with methods such as semi-empirical models which incorporate physical-insights along with data-fitting. PBLM provides a framework wherein intermediate models (IM), which capture (some) physical aspects of the problem, are incorporated into modern GLM tools to substantially improve the predictions of the latter, minimizing the reliance on costly experimental measurements or high-resolution high-fidelity numerical solutions. To illustrate the versatility and efficacy of PBLM, we present three wave-ship interaction problems: (i) at speed waterline profiles, (ii) ship motions in head seas, and (iii) threedimensional breaking bow waves. PBLM is shown to be robust and produce error rates at or below the uncertainty in the generated data, at a small fraction of the expense of high-resolution numerical predictions.

Bio-inspired pressure sensing for active yaw control of underwater vehicles

“…Another way to improve the speed of state estimation for implementation in a realtime control system is to use a physics-based learning model (PBLM) [43]. A physicsbased learning model provides a framework in which intermediate models are used to capture important physical aspects of the problem, and these aspects are incorporated into generic learning models.…”