A Threat Ranking Algorithm for Multiple Intelligent Agents in a Simulated Environment

May 1999

Proceedings of the Twelfth International Florida Artificial Intelligence Research Society Conference (FLAIRS 1999)

Volume

Proceedings of the Twelfth International Florida Artificial Intelligence Research Society Conference (FLAIRS 1999)

All Papers

It is often of utmost importance for a person to be able to determine which of several threats represent the most danger in order to react to it first. A mistake can result in damage, injury or in death. Furthermore, actions taken to address one threat, may place the person in a weaker position against other concurrent threats. This paper describes a methodology that uses heuristics to rank the various threats in order of increasing importance. The most important threat is then addressed through a proposed course of action that eliminates the threat. It then performs a simulation of the most likely outcome of the proposed course of action taken, to determine whether anything has changed with respect to the other threats. This validates the proposed course of action and permits it to be implemented. On the other hand, if the results of the simulation are such that the proposed course of action creates more problems than it solves, then it is discarded, and a new course of action is proposed. A testbed is developed and tested extensively to determine the viability of the concept.