Probabilistic Particle Filter Anchoring (PPFA): A Novel Perspective in Semantic World Modeling for Autonomous Underwater Vehicles With Acoustic and Optical Exteroceptive Sensors

Creating an accurate world model of the scenario where an autonomous underwater vehicle is navigating can be considered a crucial stage for understanding the surrounding environment. As a result, the targets detected by an automatic target recognition (ATR) architecture alongside their localized pos...

Full description

Saved in:
Bibliographic Details
Published in:IEEE journal of oceanic engineering Vol. 50; no. 2; pp. 1065 - 1086
Main Authors: Topini, Alberto, Ridolfi, Alessandro
Format: Journal Article
Language:English
Published: IEEE 01.04.2025
Subjects:
Automatic target recognition (ATR)
Autonomous underwater vehicles
autonomous underwater vehicles (AUVs)
Cameras
Inspection
marine robotics
Marine robots
Optical imaging
Optical sensors
Probabilistic logic
semantic underwater world modeling (WM)
Semantics
Sensors
Target recognition
Three-dimensional displays
ISSN:0364-9059, 1558-1691
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Creating an accurate world model of the scenario where an autonomous underwater vehicle is navigating can be considered a crucial stage for understanding the surrounding environment. As a result, the targets detected by an automatic target recognition (ATR) architecture alongside their localized positions, must be handled, selected, and filtered to get a symbolic representation of the underwater context. Even though the specific world modeling (WM) architecture may vary, current WM methodologies usually rely on the 3-D localization knowledge of the detected target by introducing a nonnegligible constraint. Motivated by the aforementioned considerations, a novel probabilistic particle filter anchoring (PPFA) approach has been developed. Starting from ATR 2-D results, the PPFA methodology aims at providing a semantic 3-D representation of the subsea environment by merging the upsides of both data association and object tracking, handled by a custom designed particle filter with resampling.
ISSN:0364-9059
1558-1691
DOI:10.1109/JOE.2024.3492537