Jeffrey Pattison

(Advisor: Prof. Dimitri N. Mavris)

will defend a doctoral thesis entitled,

UAS Mission Planning in Urban Environments to Support First Responders

On

Monday, April 8
1:00 p.m. EDT
Weber SST III, CoVE

and 

Microsoft Teams

Abstract
The past decade has seen a tremendous increase in the use of Unmanned Aerial Systems (UAS). What was once exclusively used by the military is now a critical component for a wide range of applications, including deliveries and logistics, construction, and law enforcement. Police departments around the world are beginning to see the potential UAS have for responding to emergency events. These UAS can reduce response times, be used to deescalate events, and provide crucial information to ground personnel prior to arrival. However, introducing UAS provides significant difficulties. Human operators and pilots are required to ensure safe operations and regulatory compliance. The Federal Aviation Administration has imposed strict regulations on the use of UAS in populated areas, restricting the autonomous capabilities of UAS. For UAS to be able to operate more autonomously with less human input, additional safety measures and assurance of acceptably safe operations are required. This thesis explores how to incorporate risk assessment into UAS mission planning for emergency response to introduce additional safeguards without significantly sacrificing the UAS response capability.

The major areas of research studied in this thesis include UAS risk estimation methods, UAS route planning with risk incorporated, and optimizing a system of UAS to respond to emergencies when risk is considered. Because UAS are relatively new compared to manned aircraft, UAS lack historical flight data required for risk assessment like manned aircraft. A new machine learning model is proposed that can be used for evaluating UAS risk in a more time efficient manner than the physics-based modeling and simulation methods commonly used for risk estimation. Response time is critical for emergency events, and the route a UAS takes to reach the emergency directly affects its ability to respond. This work also studies various route planning methods that can account for UAS risk to find a suitable route planning configuration that meets the demands for using UAS as a first responder. Another critical component to the response time is intelligently selecting UAS launch locations. The performance of Integer Linear Programming, Genetic Algorithms, and a hybrid algorithm are compared to determine the most suitable method for finding the optimal launch locations to minimize response time for a system of three UAS when ground risk is incorporated into the emergency response route planning. Using a software in the loop flight simulator and a vehicle simulation environment, an overarching experiment demonstrates the effectiveness of the proposed approach for incorporating risk into mission planning to see how the proposed approach impacts UAS emergency response.

Committee

·         Prof. Dimitri Mavris – School of Aerospace Engineering (advisor)

·         Dr. Michael Balchanos – School of Aerospace Engineering

·         Dr. Brian German – School of Aerospace Engineering

·         Prof. Kyriakos Vamvoudakis – School of Aerospace Engineering

·         Mike Wilson-Federal Aviation Administration