MAMBAT (Multiple-Animal Model-Based Acoustic Tracking): Improving underwater passive acoustic multi-

Abstract

Model-based localization and tracking methods provide an effective way to incorporate environmental uncertainties to estimated posit,ion uncertainty and bounds when working in propagation-dependent environments. Established localization approaches exist in cases of, single calling animals, or in cases where the call classification and association of the same call on different hydrophones can be,readily achieved. However, cases with multiple calling animals with calls that cannot be associated still pose a significant challen,ge for tracking algorithms. The problem is further complicated by false alarms and missed detections. Several localization approache,s have been proposed for such cases, but most still rely on source separation/ call association steps and are computationally demand,ing.The overall goal of the proposed research is to improve model-based passive acoustic methods for tracking multiple marine mammal,s, and to apply methods to new and challenging practical scenarios. Specifically we aim to develop and integrate a non-traditional m,ulti-target tracking framework based on random finite sets to advance multiple-animal model-based methods. We aim to automate simult,aneous tracking of multiple sources with diverse characteristics (broadband and narrowband), without requiring prior association, cl,assification or source separation steps. The framework will be computationally efficient and will incorporate missed detections, fal,se alarms, and source appearance and disappearance in the problem formulation. While the ultimate aim is to develop framework that i,s platform agnostic, method development will motivated by, tested on, and applied to large bottom mounted arrays (AUTEC and PMRF) an,d drifting 2-element vertical arrays (DASBRS).

Document Details

Document Type
DoD Grant Award
Publication Date
Oct 06, 2022
Source ID
N000142212772

Entities

People

  • Eva-Marie Nosal

Organizations

  • Office of Naval Research
  • United States Navy
  • University of HawaiĘ»i System

Tags

Readers

  • Adaptive Control and Estimation with Uncertainty in Dynamic Systems.
  • Sensor Fusion and Tracking Systems.
  • Systems Analysis and Design