Aerospace student pilots unmanned planes with an eye toward a future academic career
October 3, 2014
When Mark Palframan sets to work, he's at Virginia Tech's Kentland Research Farm. No dirt is involved, though. His hands are on a souped-up remote control device, and his eyes look to the air, set sharp on the unmanned E-SPAARO aircraft as it soars 400 hundred feet above the ground.
The Electric-SPAARO -- short for Small Platform for Autonomous Aerial Research Operations -- is a small unmanned aerial system operated by Virginia Tech's Nonlinear Systems Lab that can fly either autonomously or by remote control.
It is used for testing prototype wings and control surfaces, validating air-data sensors, and collecting and analyzing air samples. This is not a toy of play. Its mission, and that of the lab's, is not play, but forward-looking research.
A doctoral student originally from Middletown, New Jersey, Palframan arrived at Virginia Tech as a master's degree student having no prior experience with aircraft.
As an undergraduate at Lafayette College, he studied mechanical engineering with a focus on turbulent flow structures in water. There, Palframan built a highly precise water channel to carry out vortex visualization experiments.
"Being able to do cutting-edge, hands-on research was exciting, and it was at that point that I knew I wanted to do more," he said. More, though, meant change. Palframan decided to refocus his studies on control theory. After seeing the Nonlinear Systems Lab, where students could validate research on their own aerial autonomous systems that they also designed, built and tested, Palframan found he was hooked.
Palframan's role at the lab, run by Craig Woolsey and Mazen Farhood of the Department of Aerospace and Ocean Engineering, started with support roles on the flying team and taking a class to prep for the Federal Aviation Administration private pilot exam.
He trained on a small, remote-controlled T-28 Trojan plane. When the lab's designated pilot left, Palframan moved into the role, serving as E-SPAARO's main pilot. He also worked on building, maintaining, and regulating the lab's unmanned aircraft fleet.
"The E-SPAARO platform is essentially a flying pick-up truck, albeit a pretty one," Palframan said.
The vehicle may be tasked with carrying a large payload, so its design is modular with changeable wings and fuselage panels. It can carry a wide array of sensors and computational equipment for handling different in-flight experiments.
The removable wing design allows experimental, flexible wings to be easily tested, with sensors and cameras tracking shape changes in the wing as it flies. In addition to research at Virginia Tech, SPAARO has been used in research with several commercial groups, including NASK Inc., Blackbird Technologies Inc., and Aeroprobe Corp., the latter based in the Virginia Tech Corporate Research Center.
Palframan worked on his master's degree thesis under Woolsey and David Schmale III, an associate professor with the College of Agriculture and Life Sciences, who studies microbes in the atmosphere and their effect on crops and plant life.
"Mark developed a system to capture particulates from the atmosphere and inject them into a miniaturized Surface Plasmon Resonance biosensor that he integrated into our SPAARO unmanned aircraft system," said Woolsey. "This device can provide a precise measurement of the concentration of a specific biological agent -- for example, anthrax -- in a sample. As far as we know, this was the first time an SPR device had been flown and operated aboard an autonomous aircraft."
Palframan is now using E-SPAARO for his dissertation, which focuses on designing, testing, and analyzing autopilot programs for autonomous aircraft. It is hoped the programs would allow the aircraft to precisely follow a prescribed flight path while handling external disturbances, such as wind and turbulence, sensor noise, and uncertainty in the mathematical model of the aircraft.
"The idea is to cut out expensive ground testing, such as in a wind tunnel, which would be aimed at generating a highly accurate mathematical model of the aircraft to use in the generation of an autopilot controller," Palframan said. "Instead, we take a potentially less accurate mathematical model, which is much easier to obtain, and incorporated uncertainty analysis into the controller design process to create a controller that's guaranteed to be successful in the midst of disturbances and uncertainties."
Palframan said he wishes to continue working with unmanned aircraft. After gaining industry experience, he sees himself reentering academia as a professor. The education bug hit Palframan when he was teaching sophomore-level aerospace labs.
"It can be very rewarding to see the things you're teaching finally click with the students," he said.