Date of Award

3-2020

Document Type

Thesis

Degree Name

Master of Science in Systems Engineering

Department

Department of Systems Engineering and Management

First Advisor

David R. Jacques, PhD

Second Advisor

John M. Colombi, PhD

Abstract

The implementation and testing of autonomous and cooperative unmanned systems is challenging due to the inherent design complexity, infinite test spaces, and lack of autonomy specific measures. These challenges are limiting the USAF's ability to deploy and take advantage of tactical and strategic advantages offered by these systems. This research instantiates an Autonomous System Reference Architecture (ASRA) on a Wide Area Search (WAS) scenario as a test bed for rapid prototyping and evaluation of autonomous and cooperative systems. This research aims to pro- vide a framework to evaluate the system’s ability to achieve mission and autonomy objectives, develop reusable autonomous behaviors, and develop reusable cooperative decision making algorithms. For this research and application to the WAS mission, metrics of autonomy were derived from literature requirements for autonomous systems implementing reactive architectures and control: responsiveness, robustness, and perception accuracy. Autonomous behaviors, to include more complex behaviors combining simple (atomic) behaviors were developed, and a variety of cooperative decision rules were defined. The subsequent evaluation implemented a face centered cubic design of experiments over four scenarios including a single vehicle, and three levels of cooperation between two vehicles. Following a rigorous test plan, the tests were conducted in simulation implementing automated testing and expedited analysis. The test results were used to create a response surface model to characterize the system and conduct multiple response optimization to determine an optimal configuration that maximizes area searched, percent detected, and perception accuracy in a given target density.

AFIT Designator

AFIT-ENV-MS-20-M-220

Share

COinS