Road conditions

Vehicular Ad-Hoc Networks (VANETs) are essential to intelligent transportation and defence systems, where dependable communication must be achieved regardless of frequent topology changes and significant node mobility. This paper proposes the Grid-Based Structured Routing Algorithm (GBSRA), which segments vehicles into structured, grid-based clusters with dynamic cluster head selection to enhance routing stability and scalability. Simulation results confirm that GBSRA achieves a packet delivery ratio of about 90% with an average end-to-end delay of 61ms Simulation results ensures that GBSRA achieves a packet delivery ratio of about 90% with an average end-to-end delay of 61ms. It confirms its suitability for real-time vehicular communication. Then, in order to enhance predictive reliability, three machine learning (ML) models, namely Logistic Regression (LR), Support Vector Machine (SVM), and Random Forest (RF), will be applied for validating network performance. Logistic Regression gives balanced accuracy with a Receiver Operating Characteristic – Area Under the Curve (ROC-AUC) of 0.704, while Random Forest yields the highest recall for detecting the risk and SVM provides strong precision in low-risk classification. The integration of GBSRA with machine learning creates a hybrid framework that solves problems related to scalability, latency, and predictive reliability. This makes it a strong solution for safety-critical VANET applications in both civilian and military settings.