The word ‘drone’ is often associated with mindlessness.
Lately, the world has been introduced to a new type of drone — one that actually has a mind of its own. Drones, or unmanned aerial vehicles (UAVs), are aircraft that offer varying degrees of autonomy, similar to how an autopilot system in a commercial jetliner functions. However, UAVs are planes that cannot only fly and navigate, but land, take off and complete objectives with no onboard pilot and, often, no direct user guidance.
UAVs have many applications beyond their widely-publicized military use. UAVs patrol the U.S. border, help fight wildfires in California and aid in predicting seasonal flooding in the Red River Valley of North Dakota.
Over the last decade, a lab at VCU has been on the forefront of research into just how these self-sufficient flying machines navigate, see and even communicate among one another. But long before planes without pilots roamed the skies and students collaborated to build a better mousetrap, the seeds of the VCU’s Unmanned Aerial Vehicle Research Lab were sown by a young boy’s dream of becoming a Navy pilot.
Something out of nothing
Robert Klenke spent his youth with airplane models of all shapes and sizes, imagining the day he would sit in the cockpit like his father. Klenke began flying remote-controlled planes in grade school, but took his hobby further and earned his solo pilot’s license by age 15.
“I’ve been around airplanes all of my life,” he said.
In his first step to becoming a pilot, Klenke attended Virginia Military Institute, earning a bachelor’s in electrical engineering while honing his skills flying privately. He applied to Navy flight school to officially begin his career.
Behind his striped polo and neat, fatherly features, it seems easy to picture a teenage, crew-cut Klenke donning a Navy flight suit, though his humble, matter-of-fact tone is a far cry from “Top Gun”-era Tom Cruise.
“I actually got in,” he recalled. “I went down to Pensacola, but failed the eye exam.”
Klenke was disappointed, but upheld his commitment to the Navy. He oversaw construction contracts as a project manager for the Navy, continuing to fly remote-controlled planes while striving to find new ways to utilize his knowledge of aviation and computer engineering.
After completing his time in the Navy, Klenke studied at the University of Virginia for master’s, doctoral and post-doctoral work. In 1998, he joined the VCU School of Engineering which had started two years earlier.
“It was an opportunity to get in on the ground floor of something brand new, growing and exciting,” Klenke said.
“Let’s make a UAV…”
During his early years at VCU, Klenke maintained many of his consulting projects outside of VCU. He made connections at NASA Langley and in local radio-controlled hobbyist circles, continuing to build and fly model aircraft.
During the summer of 2003, Klenke’s consulting firm hired two interns from VCU. He guided the students in remote-controlled model research, where for the first time, VCU married the idea of independent data systems on model airplanes.
Klenke projected that the demand for unmanned flight systems research would expand and that VCU students could be uniquely positioned for work in the field after a summer working on research that proved a precursor to the current UAV research. Inspired, he founded VCU’s UAV Lab.
The lab contracted Klenke’s would-be side projects from government and military researchers, solidifying his passion as an official part of the VCU research community.
While working on a NASA project, Klenke caught wind of a yearly collegiate UAV competition held in Maryland by the Association for Unmanned Vehicle Systems International, a major lobbying association representing 7,500 UAV industry members.
Competing universities implemented a given flight control system to navigate a course and complete objectives, but Klenke knew students could do much more working from scratch.
Knowing that engineering students yearned for hands-on, ground-up tasks for their senior design projects, Klenke tasked himself with assembling the perfect group.
“I got (four) seniors together for their design project and said, ‘Let’s make a UAV autopilot,’” he said.
The competition’s objective: identify six military targets through a visual feed from an airborne UAV and glean information about their appearance, placement and other characteristics while flying a stable course.
The Aviation Advanced Technology Directorate at Ft. Eustis, Va. donated two military Mig-27 target drones for the VCU students to use in the competition. After adapting and testing their flight control system all year, the engineering students finished the prep on the plane by painting it black and gold.
The first-generation system read and interpreted GPS data, offering flight stabilization, data control and link-loss assurance, which are backup commands that prevent crashes if the vehicle loses contact with the operator.
Of the eight teams competing in 2004, including Cornell University, Johns Hopkins University and North Carolina State University, VCU was the only one to complete the objective. They won the first prize of $5,000 and a $500 bonus for best flight, proving that students could create a commercially competitive flight control system of their own.
The VCU UAV lab celebrated its first birthday that summer.
NASA comes calling
Beyond yearly competitions and open-source updates, the lab faced their first major technological hurdle from NASA in 2005.
The VCU UAV lab was contacted by Project AirSTAR, where Klenke had previously trained RC pilots working with a scale model of a Boeing 757. Researchers used Wallops Island, Va. as a site to test flight safety conditions for commercial airlines using the models. Called the Generic Transport Model, the planes were outfitted with jet-turbine engines — not exactly the remote-controlled planes that hobbyists toy around with in their garages.
No commercially available autopilot existed for jet-turbine craft, but after two years of master’s thesis updates and senior design projects, Klenke’s lab was ready to adapt their flight control system to meet NASA specifications.
VCU’s autopilot guided the model through repeated flight paths impossible for a human pilot to fly, refining NASA’s research by eliminating extraneous research variables. The system also provided the same link-loss assurance protocols that the first UAV competition team had built into their plane, preventing crashes during testing and saving researchers time and money.
Some Virginians fear government use of UAVs, especially in respect to potential invasions of privacy.
While Klenke assured that members of the UAV community are sensitive to privacy issues, people like VCU philosophy major James Humphreys fears the fabled Gorgon Stare of current UAVs: Their ability to scan a 4 km by 4 km square with facial recognition software.
“Potential benefits are hard to ignore, but the real problems stem from human nature and prior government transgressions against U.S. citizens,” Humphreys said, referencing incidents such as the killing of three Americans in two drone strikes in Yemen in the fall of 2011. “For UAVs to be used effectively and in the public’s confidence, those in law enforcement must strictly adhere to the ‘protect and serve’ order.”
Virginia legislators, with help from the ACLU, hope to curb potential infringement of privacy by any government body in Virginia using UAVs with House Bill 1616, the first legislation limiting UAV use in Virginia’s history. The bill is sponsored by Delegate Todd Gilbert (R-15).
While provisions specifically seek to limit the use of UAVs for gathering information and how that information can be disclosed, the bill also provides that universities should maintain the right to use unmanned vehicles in academic or research applications.
“By working with UAVs, you see their tremendous potential … (for) anything dirty, dull or dangerous, ” Klenke said.
VCU electrical engineering junior Ben Young joined the UAV competition team in the spring of 2011. He immediately felt a difference between the UAV lab and other labs at VCU.
“The idea here is that undergraduates and graduates work together, so there is always a lot to learn and problems of different magnitudes can be solved,” Young said.
Young came to VCU after plans fell through with N.C. State, dashing his hopes of working with UAVs in his home state. He eventually found the VCU competition team while searching the web.
“The cool thing about working with UAVs is that no matter what your area of interest is, it has something to offer,” Young said.
Recently, research into flight control systems at the lab has branched in two different directions because of the varying applications of UAVs. One line of research focuses on developing a system for complex aircraft, currently used for launching gliders from a flying UAV for the Navy. The other line of research is into simplistic systems meant for coding UAV-to-UAV interaction and autonomous decision-making, the first step in collision-avoidance technology necessary for commercial airspace integration.
Master’s researcher Siva Tejapatibandla currently studies UAV-to-UAV interaction.
“For UAVs to collaborate and solve a problem, there needs to be a robust and reliable data link between them,” Tejapatibandla said. “I’m working to develop that link between moving UAVs.”
Tejapatibandla has already received multiple job offers in advance of his May 2013 graduation based on his research.
Tim Bakker, the lab’s only Ph.D. researcher, works on the RAMS simulator, software built from scratch by the lab to model complex algorithms that drive UAV-to-UAV interaction.
“The simulator project is an ongoing process, and we hope to have it made public as an open-source project to other academic institutions,” Bakker said.
Bakker was intrigued by the lab during his first visit to VCU and was immediately interested in joining after speaking to Klenke. With more than two years of research complete, he said he feels that hands-on learning, collaborative environments and the diverse group of students have made him “a better engineer.”
“It was great to see such a unique topic being covered, which, for many schools, is not financially or otherwise academically possible,” Bakker said.