Intelligent Machine Learning Framework for Classification of Vehicular Traffic in Internet of Vehicles

In a world that is always changing due to population growth, it is difficult to forecast how routes will behave in the future, especially with regard to traffic. In this work, Traci was used to generate traffic-related data in a real-world setting, including waiting periods, coordinates, speed, and displacement. After that, cars were categorized according to how long they took to wait, and performance was examined using a variety of machine learning methods, such as Random Forest (RF), K-Nearest Neighbor (KNN), Support Vector Machine (SVM), and K-means. The RF algorithm fared better than other algorithms, according to empirical studies, in estimating waiting times and evaluating traffic conditions. After that, the study used the Rapidly Exploring Random Tree (RRT) algorithm, a path-planning and simplification technique, to find vehicle travel paths that would not cause collisions. By decreasing wait times and improving customer satisfaction, this strategy aims to optimize journey routes and cut related expenses. Drones are used in the proposed system, and base stations and ad hoc networks work together to improve overall performance.