Date of Award

Spring 1-1-2011

Document Type

Dissertation

Degree Name

Doctor of Philosophy (PhD)

Department

Aerospace Engineering Sciences

First Advisor

Eric W. Frew

Second Advisor

Brian M. Argrow

Third Advisor

Adam Houston

Abstract

This work describes the development of a network-centric unmanned aircraft system (UAS) for in situ sampling of supercell thunderstorms. UAS have been identified as a well-suited platform for meteorological observations given their portability, endurance, and ability to mitigate atmospheric disturbances. They represent a unique tool for performing targeted sampling in regions of a supercell thunderstorm previously unreachable through other methods.

Doppler radar can provide unique measurements of the wind field in and around supercell thunderstorms. In order to exploit this capability, a planner was developed that can optimize ingress trajectories for severe storm penetration. The resulting trajectories were examined to determine the feasibility of such a mission, and to optimize ingress in terms of flight time and exposure to precipitation.

A network-centric architecture was developed to handle the large amount of distributed data produced during a storm sampling mission. Creation of this architecture was performed through a bottom-up design approach which reflects and enhances the interplay between networked communication and autonomous aircraft operation. The advantages of the approach are demonstrated through several field and hardware-in-the-loop experiments containing different hardware, networking protocols, and objectives.

Results are provided from field experiments involving the resulting network-centric architecture. An airmass boundary was sampled in the Collaborative Colorado Nebraska Unmanned Aircraft Experiment (CoCoNUE). Utilizing lessons learned from CoCoNUE, a new concept of operations (CONOPS) and UAS were developed to perform in situ sampling of supercell thunderstorms. Deployment during the Verification of the Origins of Rotation in Tornadoes Experiment 2 (VOR- TEX2) resulted in the first ever sampling of the airmass associated with the rear flank downdraft of a tornadic supercell thunderstorm by a UAS.

Hardware-in-the-loop simulation capability was added to the UAS to enable further assessment of the system and CONOPS. The simulation combines a full six degree-of-freedom aircraft dynamic model with wind and precipitation data from simulations of severe convective storms. Interfaces were written to involve as much of the system's field hardware as possible, including the creation of a simulated radar product server. A variety of simulations were conducted to evaluate different aspects of the CONOPS used for the 2010 VORTEX2 field campaign.

Share

COinS