Route Optimisation with OSM & AI

The growing need for intelligent and effective transportation in expanding urban areas has required the creation of intelligent route planning systems. This project here has a complete and up-to-date solution with the name “Route Optimization with OpenStreetMap (OSM) using titled “Route Optimization with OpenStreetMap (OSM) using Real-Time IoT GPS, XGBoost Machine Learning, SHAP Explainable AI, and AR Navigation through Supabase and Flutter.” The main focus of this project is to dynamically optimize routes using live GPS data fetched using an ESP32 microcontroller combined with a GPS module. This data is streamed to a FastAPI backend and saved in Supabase, a cloud-based backend service that offers real-time updates and storage. Stored GPS data is further processed through an XGBoost classification model to forecast the most appropriate travel mode-e.g., bike, car, train, or flight-on the basis of current location, travel patterns, and history of context. Besides prediction, clustering algorithms are employed to cluster patterns of routes on the basis of traffic density, region, and travel behavior. OpenStreetMap (OSM) APIs are used for fetching and analysis of route data by the project and it follows shortest path algorithms for best travel choices. Additionally in the manner of travel mode prediction, adding credibility and trustworthiness to the AI choices. The Flutter mobile app is the frontend interface, providing users with real-time AR-based navigation and choice of start and destination points. AR features are superimposed through sophisticated Flutter widgets to aid navigation and situational awareness. Real-world practicability is emphasized in the implementation utilizing low-cost hardware and free opensource software stacks. In contrast to conventional GPS navigation systems, this system is designed to be able to realtime adapt to evolving road conditions, user needs, and environmental trends. Widespread testing verifies that the system provides reliable predictions, efficient pathfinding, and easy-to-use navigation with minimal latency.