Near-nash Targeting Strategies for Heterogeneous Teams of Autonomous Combat Vehicles

The Nash equilibrium concept of non-zero sum games is one possible option available to military planners seeking strategies to control large numbers of autonomous assets operating in an adversarial environment. To implement the Nash strategies inherently necessitates making assumptions on possible adversarial actions. However, the Nash concept suffers from one major difficulty which limits its potential usefulness. A Nash equilibrium may not always exist in pure strategies. In this paper we introduce the concept of Near-Nash strategies as a mechanism to overcome this difficulty. We then illustrate this concept by deriving the Near-Nash strategies for a military game where a unique Nash is not guaranteed to exist. We use these strategies as the basis for an intelligent battle plan for heterogeneous teams of autonomous combat air vehicles in the Multi-Team Dynamic Weapon Target Assignment model.