Artificial Intelligence/Machine Learning (AI/ML) Battle Management for Rapid, Interactive Weapon Selection

Threat missile systems are evolving in type, volume, and capabilities. Current missile defense programs rely on Kinetic (K) kill weapons and various sensors such as ground-based radars, sea-based radars, and Electro-Optical and Infrared (EO/IR) sensors on aircraft and spacecraft to defend against the evolving threats. Kinetic weapons are limited by available interceptor numbers, engagement volume, and time to intercept potentially limiting the CommanderÆs battle plan options. To increase our defense capabilities, Non-Kinetic (NK) weapons, also known as Directed Energy (DE), will be integrated into the Missile Defense System (MDS). These new NK weapons will bring game changing speed-of-light kill capabilities to the MDS. While NK weapons are game changers, they are not without their own limitations. Weather conditions can degrade NK weapon effectiveness which must be a consideration for the MDS and Commanders in planning and executing engagements. To address this complex battlespace, the Commander will need new tools for situational awareness, weapon selection, and rapid decision aides for real-time assessment and execution. ExoAnalytic Solutions (ExoAnalytic) is proposing an integrated Artificial Intelligence (AI)/Machine Learning (ML) approach that blends the speed and optimization advantages of deep reinforcement learning with the human understandable advantages of probabilistic-based reasoning. The system will be fast and accurate but include the human in the loop in a telegraphic and rapid-feedback manner allowing commanders the opportunity to rapidly assess multiple battle plans updated over time as the battle evolves. Approved for Public Release | 24-MDA-11906 (16 Sep 24)