PROPHETEER: Predictive Planning for Real World Adversarial Domains

Period of Performance: 03/19/1998 - 12/19/1998

$100K

Phase 1 SBIR

Recipient Firm

Stottler Henke Associates
1650 South Amphlett Boulevard, Suite 300
San Mateo, CA 94402
Principal Investigator

Research Topics

Abstract

We propose an innovative combination of artificial intelligence (AI) techniques in the design of a predictive planning and preemption (PPP) system. Specifically, we propose to enhance military planning processes by exploiting predictive battlespace knowledge to shape the future actions of the enemy to the benefit of friendly forces. By drawing on our experience in planning, machine learning, and behavior modeling and recognition, we have devised an eclectic approach to predictive planning that is applicable to real world military domains. We propose to integrate recent advances in plan recognition and data mining with modern planning and execution systems to form a unique means for proactive and dynamic warfare planning. Additionally, we will incorporate mobility, communications, logistics, information warfare, and other non hard kill weapons based operations into the planning process. The tight integration of simulation and modeling capabilities with this extended planning process will greatly improve the ability of the US military to control the pace and phasing of military operations and substantially decrease the risk of surprise. We will prove the feasibility of our approach by developing and demonstrating a proof-of-concept prototype in Phase I.