Design and Implementation of Adaptive Game AI using Priority Matrices and Vector-Based Navigation

In modern gaming, the realism and responsiveness of non-player characters (NPCs) significantly enhance player immersion and engagement. This paper proposes an adaptive. AI framework for NPCs using priority matrices for behavior selection and vector-based navigation for pathfinding and movement. The priority matrix evaluates environmental and internal game parameters to assign dynamic action priorities. Simultaneously, vector-based navigation enables fluid, real-time navigation in complex environments by computing motion vectors influenced by goal attraction, obstacle avoidance, and inertia. The hybrid framework was implemented in a custom game simulation built with Unity and C, demonstrating improved NPC adaptability and naturalistic behavior. Comparative evaluation shows our model enhances responsiveness and decision quality while maintaining computational efficiency. This paper details the design, implementation, and validation of the system, offering a novel and scalable approach to adaptive game AI.