A Target Classification Decision Aid

TL;DR

This paper introduces Horizon, a prototype decision aid that uses evidential reasoning to help submarine sonar teams classify targets more effectively amid uncertain and abundant information.

Contribution

It presents a novel evidential reasoning-based software tool designed to fuse imprecise data for submarine target classification.

Findings

Horizon effectively fuses uncertain information for target classification.

The software prototype demonstrates potential for operational use.

It addresses information overload in sonar target analysis.

Abstract

A submarine's sonar team is responsible for detecting, localising and classifying targets using information provided by the platform's sensor suite. The information used to make these assessments is typically uncertain and/or incomplete and is likely to require a measure of confidence in its reliability. Moreover, improvements in sensor and communication technology are resulting in increased amounts of on-platform and off-platform information available for evaluation. This proliferation of imprecise information increases the risk of overwhelming the operator. To assist the task of localisation and classification a concept demonstration decision aid (Horizon), based on evidential reasoning, has been developed. Horizon is an information fusion software package for representing and fusing imprecise information about the state of the world, expressed across suitable frames of reference. The Horizon software is currently at prototype stage.