adaptive_q_learning_vnet_qos_rs_optimization

Vehicular Fog Computing (VFC) is a cornerstone of next-generation Intelligent Transportation Systems (ITS), providing distributed, low-latency computational resources for connected vehicles. However, ensuring dynamic responsiveness to safety-critical events such as ambulance movement, crash alerts, and road hazards, while maintaining energy-efficient and trustworthy communications, remains an open challenge. This research introduces an augmented reinforcement learning framework that leverages a novel Augmented Priority-Entropy Reward Function (APERF), integrating normalized multi-metric QoS, exponential event-driven priority scaling, energy-awareness, and policy entropy. The model enables a Q-learning agent to dynamically optimize resource allocation and incentive mechanisms under complex, event-driven urban traffic conditions. Through extensive simulation on synthetic VFC datasets, this approach outperforms classical and trust-based baselines, achieving robust event-prioritized Quality of Service (QoS), significant energy savings, and enhanced policy adaptability. I provide ablation studies, comparative metrics, and insights for real-world deployment in future ITS environments.