Mismatch-Robust Underwater Acoustic Localization Using A Differentiable Modular Forward Model

Dariush Kari, Yongjie Zhuang, Andrew C. Singer

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

In this paper, we study the underwater acoustic localization in the presence of environmental mismatch. Especially, we exploit a pre-trained neural network for the acoustic wave propagation in a gradient-based optimization framework to estimate the source location. To alleviate the effect of mismatch between the training data and the test data, we simultaneously optimize over the network weights at the inference time, and provide conditions under which this method is effective. Moreover, we introduce a physics-inspired modularity in the forward model that enables us to learn the path lengths of the multipath structure in an end-to-end training manner without access to the specific path labels. We investigate the validity of the assumptions in a simple yet illustrative environment model.

Original languageEnglish (US)
Title of host publication2025 59th Annual Conference on Information Sciences and Systems, CISS 2025
PublisherInstitute of Electrical and Electronics Engineers Inc.
ISBN (Electronic)9798331513269
DOIs
StatePublished - 2025
Event59th Annual Conference on Information Sciences and Systems, CISS 2025 - Baltimore, United States
Duration: Mar 19 2025Mar 21 2025

Publication series

Name2025 59th Annual Conference on Information Sciences and Systems, CISS 2025

Conference

Conference59th Annual Conference on Information Sciences and Systems, CISS 2025
Country/TerritoryUnited States
CityBaltimore
Period3/19/253/21/25

Keywords

  • few-shot adaptation
  • forward modeling
  • mismatch
  • physics-inspired modeling
  • test time adaptation
  • underwater acoustic

ASJC Scopus subject areas

  • Computer Science Applications
  • Information Systems
  • Control and Optimization
  • Artificial Intelligence
  • Signal Processing
  • Modeling and Simulation

Fingerprint

Dive into the research topics of 'Mismatch-Robust Underwater Acoustic Localization Using A Differentiable Modular Forward Model'. Together they form a unique fingerprint.

Cite this