Physics perception in sloshing scenes with guaranteed thermodynamic consistency

TL;DR

This paper introduces a neural network-based method for reconstructing and predicting the full state of sloshing liquids from limited surface measurements, ensuring thermodynamic consistency and enabling real-time fluid reasoning.

Contribution

It presents a novel approach combining RNNs and reduced-order modeling to achieve physically consistent fluid predictions from partial data.

Findings

Successfully reconstructs full fluid states from surface measurements.

Ensures thermodynamic principles are satisfied in predictions.

Operates in real-time with integrated computer vision system.

Abstract

Physics perception very often faces the problem that only limited data or partial measurements on the scene are available. In this work, we propose a strategy to learn the full state of sloshing liquids from measurements of the free surface. Our approach is based on recurrent neural networks (RNN) that project the limited information available to a reduced-order manifold so as to not only reconstruct the unknown information, but also to be capable of performing fluid reasoning about future scenarios in real time. To obtain physically consistent predictions, we train deep neural networks on the reduced-order manifold that, through the employ of inductive biases, ensure the fulfillment of the principles of thermodynamics. RNNs learn from history the required hidden information to correlate the limited information with the latent space where the simulation occurs. Finally, a decoder returns data back to the high-dimensional manifold, so as to provide the user with insightful information in the form of augmented reality. This algorithm is connected to a computer vision system to test the performance of the proposed methodology with real information, resulting in a system capable of understanding and predicting future states of the observed fluid in real-time.