Agent-Based Mixed-Initiative Collaboration

Agent-Based Mixed-Initiative Collaboration
A Research Project under the DAGSI/AFRL Aerospace Research Program

Final Report (WSU-CS-02-01)


In the most basic sense, this research intends to examine the nature of human collaboration with the goal of gaining a better understanding of the cognitive processes involved. In the most practical sense, the research is to build systems that facilitate human collaboration and to model on computers the behavior that leads to successful cooperation. To study collaboration, we will study a particular type of collaboration that is relevant to many intelligent activities; that is, we will examine planning by teams of people that are engaged with computer systems and vast amounts of information. One of the most difficult tasks during such activity is to present to each person just the right amount of information in the form best suited to their particular piece of the team effort. To provide solutions to this problem, we will experiment with mixed-initiative approaches that synthesize human and machine decisions, case-based approaches that reuse past experience, agent-based technologies that encapsulate specialized knowledge, and cognitive systems engineering methods of user-modeling.

Goals and Objectives

The overall project goal is to investigate agent-based and case-based frameworks for distributed collaborative planning using a goal-centered approach. The focus of this proposal is on three main research objectives associated with the main project goal:

The research will deliver a prototype computer system and user-interface that supports intelligent collaboration.

Major Participants