Improving Air Battle Management Target Assignment Processes via Approximate Dynamic Programming

Military air battle managers face many challenges when directing operations in quickly evolving combat scenarios. These scenarios require rapid decisions to engage moving and unpredictable targets. In defensive operations, the success of a sequence of air battle management decisions is reflected by the friendly force\u27s ability to maintain air superiority by defending friendly assets. We develop a Markov decision process (MDP) model of the air battle management (ABM)problem, wherein a set of unmanned combat aerial vehicles (UCAV) is tasked to defend a central asset from cruise missiles that arrive stochastically over time