Modelling Aircraft Disturbance Fields for Magnetic Navigation Using Dense ANNs and the Novel MANNTL Architecture

reportActive / Technical Report | Accesssion Number: AD1132393 | Open PDF

Abstract:

The ability to use GPS for navigation is becoming increasingly limited in certain areas of the world. Knowing this, the Air Force Research Labs is constantly looking for ways to improve alternate navigation methods such as magnetic navigation. In the interest of making advancements in aircraft disturbance xC;field modelling, Lieutenant Emery recreates models from previous works to prove results. Lieutenant Emery also introduces a novel model architecture that attempts to mix the xC;filtering properties of Tolles-Lawson with the non-linear capabilities of an artifixC;cial neural network. The introduction of this model could present better aircraft disturbance xC;field modelling and in turn, more reliable magnetic navigation in regions where GPS is not available.

Author(s):
Emery, Kyle A
Author Organization(s):
AIR FORCE INSTITUTE OF TECHNOLOGY WRIGHT-PATTERSON AFB OH
Funding Organization(s):
Office of Naval Research, Arlington , VA
Document Type:
Technical Report/Master's Thesis
Publication Date:
2021 Mar 01
Pagination:
80

Security Markings

DOCUMENT & CONTEXTUAL SUMMARY

Distribution Code:
A - Approved For Public Release
Distribution Statement: Public Release

RECORD

Collection: TRECMS
Subject Terms
Modernization Areas:
AI and Machine Learning
Communities of Interest:
Air Platforms
Descriptor(s):
neural networks, air force, artificial intelligence software, navigation, unmanned aerial vehicles, machine learning, global positioning systems, measurement, military research, supervised machine learning, recurrent neural networks, information science, magnetic fields, magnetic navigation, united states, aircrafts, convolutional neural networks
Keyword(s):
Aircraft Disturbance Fields, Magnetic Navigation, dense ann, ann (artificial neural networks), tl (tolles-lawson), tl algorithm, manntl (Mixing ann and tl), atl (augmented tl)
Subject Categories:
Navigation, Detection and Countermeasures; Mathematical and Computer Sciences
Creation Date:
2021 Jun 02
Update Date:
2021 Jul 26