Dr. Shelley Zhang on team developing project for U.S. Air Force

Dr. Xiaoquin 'Shelley' Zhang of the Computer and Information Science Department is among a team of researchers, united by Lockheed Martin Advanced Technology Laboratories, that has won a $22 million, 48-month contract from the Defense Advanced Research Projects Agency.

The group, with members from 11 universities and institutes, will develop the “Generalized Integrated Learning Architecture” to help U.S. Air Force planners leverage the skills of expert operators; the project aims for improved control of air space over battlefields and the transfer of expert  knowledge to inexperienced personnel. UMass Dartmouth’s portion of the funding totals $290,000.
 
Zhang, UMass Dartmouth’s only researcher on the team, participates in weekly telephone meetings with other members and occasionally goes to Lockheed Martin’s Shrewsbury facility to work on the project. Lockheed Martin, headquartered in Bethesda, Maryland, employs approximately 135,000 people worldwide.

In addition to Lockheed Martin ATL and UMass Dartmouth, Zhang’s other collaborators include Lockheed Martin Integrated Systems and Solutions, Lockheed Martin Advanced Technology Center, University of Maryland, Georgia Institute of Technology, Georgia Tech Research Institute, University of Illinois, Arizona State University, Stanford University, University of Wyoming, Oregon State University, and Fujitsu Laboratories of America.

Zhang said that her particular task is helping to develop the meta-reasoning executive component, which is responsible for the coordinations among several integrated learners and other reasoning components in the system. Noting that this grant is her first major external award, Zhang said, “It is very exciting for me to be part of such an important project. The biggest challenge for me will be to perform research in a large team setting that requires efficient communication and collaboration.”

Lockheed Martin plans to use advanced machine learning and research provided by the teammates, and integrate the results into tools for military planners such as “WEBAD”—a web-based tool for airspace deconfliction. Planners in air operations centers use an Airspace Control Order to define and control four-dimensional, time-space volumes in which a number of manned and unmanned aerial vehicles and weapons remain separated or deconflicted.

According to officials, the project will help create the Airspace Control Order by automatically learning the planner’s tasks from an expert. Eventually, it will outperform the new human planner by 125 percent while giving the inexperienced user an embedded, accelerated training capability. “This will revolutionize the way expert skill and knowledge are captured and transferred,” said Ken Whitebread, the project’s principal investigator at Lockheed Martin ATL in a written statement.

James Hendler, professor and director of the Joint Institute for Knowledge Discovery at University of Maryland, said, “We have a team with some of the top researchers in the country coupled with Lockheed Martin’s research integration expertise.”

At UMass Dartmouth, Zhang is in her fifth year of teaching in areas that include advanced artificial intelligence, multi-agent systems, object-oriented programming, and understanding E-commerce technologies. She received her Ph.D. and master’s degrees in computer science from UMass Amherst, and her undergraduate degree from the University of Science & Technology of China.

“I appreciate UMass Dartmouth’s College of Engineering because of the small class sizes that allow for personal interaction with students that would be impossible in a room with 200,” Zhang said. “That is very important for student achievement and it’s attractive to me as an instructor.”