Distributed Planning in a Mixed-Initiative Environment
Abstract
The USAF Command and Control (C2) is undergoing a transformation to enable a full-spectrum, joint warfighting capability. To be able to meet the future challenge of employing forces anywhere in the world in support of national security objectives, the USAF requires a highly synchronized, distributed planning and replanning capability that is flexible to adapt to any level of conflict. This paper describes an in-house program underway at the USAF Research Laboratory Information Directorate that is developing technologies to support the concepts of Network-Centric Operations (NCO). The research focus of this program is on the concepts and architecture needed to support the distributed, mixed-initiative planning required for NCO. Our system builds upon distributed blackboards and multi-agent systems to provide automated opportunistic planning capabilities for distributed C2 operations. An extensible UML model of plans has also been developed to support human-machine dialog for mix-initiative planning. The plan representation is object oriented, recursive, and supports plan fragment operations, a key concept for distributed planning. This paper will also discuss our future research directions, including the encoding of human-planner experience and expertise for rapid formation of distributed expert planning teams.
Document Details
- Document Type
- Technical Report
- Publication Date
- Jun 01, 2008
- Accession Number
- ADA486896
Entities
People
- Chad Destefano
- Joseph Carozzoni
- Kurt Lachevet
Organizations
- Rome Laboratory