Date of Award

6-14-2018

Document Type

Thesis

Degree Name

Master of Science in Electrical Engineering

Department

Department of Electrical and Computer Engineering

First Advisor

Jason M. Bindewald, PhD.

Abstract

Autonomous systems are increasingly being used for complex tasks in dynamic environments. Robust automation needs to be able to establish its current goal and determine when the goal has changed. In human-machine teams autonomous goal detection is an important component of maintaining shared situational awareness between both parties. This research investigates how different categories of goals affect autonomous change detection in a dynamic environment. In order to accomplish this goal, a set of autonomous agents were developed to perform within an environment with multiple possible goals. The agents perform the environmental task while monitoring for goal changes. The experiment tests the agents over a range of goal changes to determine how detection performance is affected by the different categories of goals. Results show that detection is highly dependent on what goal is being switch to and from. The point similarity between goals is the most significant factor in evaluating the change detection time. An additional experiment improved upon the goal agent and demonstrated the importance of having the proper perception mechanics for feedback within the environment.

AFIT Designator

AFIT-ENG-MS-18-J-003

DTIC Accession Number

AD1056649

Download

Share

COinS