Skip to content. Skip to main navigation.

UTA researcher earns grant to investigate metrics that guarantee performance of unmanned aerial vehicles in groups

Thursday, April 5, 2018 •

computer scienceengineeringresearchAll Topics

When an aircraft is built, manufacturers guarantee that it will meet certain conventional margins for performance, such as speed, stability and handling under various conditions.

Kamesh Subbarao

Kamesh Subbarao, a UTA associate professor in the Mechanical and Aerospace Engineering Department, received a grant to help multicple unmanned vehicles work in concert.

Conventional margins are standards the industry has endorsed. They are specifications that can be used to predict the stability and performance of a single vehicle with reasonable reliability. However, no such specifications exist to predict how groups of vehicles will perform in tandem.

Kamesh Subbarao, an associate professor in the Mechanical and Aerospace Engineering Department at The University of Texas at Arlington, is using a four-year, $795,427 Basic Research Challenge grant from the Office of Naval Research to create metrics to evaluate the performance of multiple unmanned vehicles that operators will be able to use to predict vehicle behavior under mission-specific conditions.

Once Subbarao has determined those metrics, he will develop a framework to identify key system parameters that will provide guaranteed performance of a group of aircraft.

Subbarao, an aerospace engineer, will focus on fully autonomous unmanned aerial vehicles, but his findings could easily be applied to land- or water-based vehicles or groups of multiple types of unmanned vehicles.

“We know how to assess the performance of a single vehicle, but not how teaming affects multiple vehicles, so I will work to identify key parameters that affect overall performance and then make recommendations on how they need to be applied,” Subbarao said.

“There are many variables that show up in a group that aren’t there in individual tasks, so if you don’t know how sensitivity to those variables will affect each vehicle, it is difficult to predict how each one will perform in a group.”

Conventional margins have been used for decades based on mathematical models of a vehicle and can be quickly evaluated to predict performance. The challenge in applying conventional margins to groups of aircraft is that each vehicle is different and each mission might call for multiple tasks within the mission. The timing and coordination of those tasks creates conditions that the creators of the performance specifications might not have considered.

For instance, if a group of unmanned aerial vehicles was sent to survey an area and each had a different sensor for a different type of data, the order in which the vehicles passed over the area could make a difference, and each vehicle would have to understand when the previous vehicle had completed its part of the mission. Coordinating airborne and ground-based vehicles on a mission also would be challenging based on how terrain or weather affected any of the vehicles involved.

“There must be a continuous exchange of information, and it is important to create parameters in such a way that mission planners will have a good idea of the overall effectiveness when they decide how to deploy resources,” Subbarao said.

Subbarao’s work is an example of how UTA researchers are supporting data-driven discovery, one of four themes of the University’s Strategic Plan 2020: Bold Solutions | Global Impact, says Erian Armanios, chair of the Mechanical and Aerospace Engineering Department.

“This is the power of building squads of autonomous aircraft to accomplish a broad range of missions efficiently,” Armanios said. “Dr. Subbarao’s unique expertise in this project is bound to achieve breakthroughs in customizing formation patterns to a given operation and allow mission planners to make informed, data-based decisions that will ensure that assets are used in the most effective way possible.”

Subbarao previously earned a grant from the Air Force Research Laboratory to improve the ability of unmanned vehicle systems to work together and better analyze data collected within an environment.

Also in the College of Engineering, Yan Wan, an associate professor of electrical engineering, is using a $998,803 National Science Foundation grant to create a networked aerial computing system would allow operators to download fused information from UAV networks and respond in real time, and would increase the vehicles’ ability to share information with each other for safer control in flight.

Electrical Engineering Professor Frank Lewis and his collaborators developed and patented a process called Integral Reinforcement Learning, by which a device learns and makes control decisions in reaction to a set of variables that changes based on each previous decision, continuously in real time, online, allowing greater autonomy and faster response.

The College offers graduate and undergraduate certificate programs in Unmanned Vehicle Systems and has several faculty members conducting research in command and control, vehicle design and other related areas. The UTA Research Institute has an FAA Certificate of Authorization to fly in the airspace around its campus, making it a viable venue for testing.

In addition, four UTARI/UTA faculty members play key roles on the Texas team that develops safety systems for unmanned aircraft at the Lone Star Unmanned Aerial Systems Center. Along with UTARI, Texas A&M Corpus Christi, Texas A&M Engineering Experiment Station, Camber Corp. of Huntsville, Ala., and the Southwest Research Institute in San Antonio were named by the Federal Aviation Administration as part of the agency’s efforts to get UAVs airborne.