Using Contexts Competition to Model Tactical Human Behavior in a Simulation

Abstract

This article describes an innovative approach for representing tactical decision-making in Autonomous Intelligent Platforms, or AIP. The Competing Context Concept, is associated with the Context-Based Reasoning (CxBR) modeling paradigm, and represents an improvement thereof. CxBR uses as its basis an intuitive structure called a Context. One specific context is always in control of the AIP, and it contains all the information required to control that AIP when in that situation. When the situation changes, a new context must be found that properly addresses the new situation. Upon finding such a new context, it becomes activated, and the old context deactivates itself. An AIP, therefore, can be controlled intelligently through a sequence of transitions among various (pre-existing) contexts. The Competing Context Concept introduces a means to control the transition process without the need to predetermine the next contexts. This paper briefly describes the CxBR paradigm and the competing context concept extension. Results from prototype testing will also be discussed.