The agent software called CAST (Collaborative Agents for Simulating Teamwork) highlights relevant data. This helps improve a team's decision-making process as well as enhances members' collaboration.

"This version of CAST provides support for teams by anticipating what information team members will need, finding commonalities in the available information and determining how that information should be processed," said John Yen, professor of information sciences and technology. "Decision making is made easier because the software offers only relevant data."

CAST was originally developed by a team of researchers at Texas A&M where Yen was a key figure. Now a faculty member in Penn State's School of Information Sciences and Technology (IST), Yen heads the Research Laboratory for Team-based Agents at the University while continuing to collaborate with Richard Volz and Michael Miller, from Texas A&M, on the software.

Initially, CAST was developed to facilitate or train human teams in the best ways to collaborate on and perform certain tasks. The research has been funded through a Department of Defense MURI (Multidisciplinary Research Program of the University Research Initiative) grant to Texas A&M, Wright State University and Penn State.

With this research, the research team is taking smart software into a new direction involving what he calls "shared mental models" to support team activities or train teams. These can include shared team goals, shared assumptions about the problem, and shared knowledge about the team structure and process.

"The inspiration came from psychologists studying the behavior of human teams who were required to process incoming information under the pressure of time constraints," Yen said.

Without being directed, members of higher-performing teams were able to provide each other with needed information. This enabled more timely and better decisions, he added. CAST does this, too.

"The more time-critical the environment in which a team operates, the more effectively it needs to process information," Yen said. "A computer program that acts as a team member may be more efficient in processing information than a human teammate."

The Penn State researcher and his collaborators see CAST as a promising technology for supporting military officers who receive from ground sensors and satellites as many as 600,000 reports every hour. Without the right information, the wrong decision can be made in the battle space, Yen said.

The software, which can be customized, also can help officers adapt more quickly to changing battlefield conditions. CAST also could be used to track potential terrorist threats or infectious diseases - any domain where information needs to be exchanged quickly or commonalities found among different cases, Yen said.

Yen had been scheduled to present this research as a keynote speaker at the second International Conference on Active Media Technology in the People's Republic of China, May 29-31.

The conference was canceled due to SARS. The paper, "On Modeling and Simulating Agent Teamwork in CAST," appears in the conference's proceedings released by World Scientific Publishing Company.

The authors from Penn State are Yen; Xiacong Fan, a postdoctoral scholar; and IST doctoral students Shuang Sun, Ray Wang and Cong Chen. Kaivan Kamali is a doctoral student in Computer Science and Engineering. Volz and Miller, Texas A&M, also were co-authors.

Related Links
Penn State
SpaceDaily
Search SpaceDaily
Subscribe To SpaceDaily Express

Thanks for being here; We need your help. The SpaceDaily news network continues to grow but revenues have never been harder to maintain. With the rise of Ad Blockers, and Facebook - our traditional revenue sources via quality network advertising continues to decline. And unlike so many other news sites, we don't have a paywall - with those annoying usernames and passwords. Our news coverage takes time and effort to publish 365 days a year. If you find our news sites informative and useful then please consider becoming a regular supporter or for now make a one off contribution.
SpaceDaily Contributor $5 Billed Once credit card or paypal		SpaceDaily Monthly Supporter $5 Billed Monthly paypal only