Date of Award

3-2021

Document Type

Thesis

Degree Name

Master of Science

Department

Department of Electrical and Computer Engineering

First Advisor

Brett J. Borghetti, PhD

Abstract

A limitation in RPA ISR operations is loss of target track if the command link is severed. For an RPA to effectively execute the ISR mission without a command link, it needs the capability to F2T targets autonomously. Automated Find Fix and Track (AFFTRAC) was developed to help solve this problem by demonstrating a proof of concept tactical autopilot. Monocular stereo vision was used to process sequential images acquired during orbit to produce a partial structural point cloud of the original structure. This partial structural point cloud was then exploited to create a holding area density for the aircraft to stay within. A simple greedy algorithm exploited this holding area density to produce aircraft turn commands to approximate tactical ISR holding. The result was that imagery from existing MQ-9 sensors was used to provide command guidance to autonomously to maintain line of sight to a target. Overall, AFFTRAC is a promising initial framework for a tactical autopilot, but additional development is needed to mature component algorithms.

AFIT Designator

AFIT-ENG-MS-21-M-069

DTIC Accession Number

AD1134239

Share

COinS