BETA 2.0
Release: 2025
AI Integration & Real-Time Data Mining
Migration to a microservices architecture (Unity, Node.js, Python) to enable advanced analytics and machine learning pipelines.
- Weather Replay: Integration of historical INMET data to drive visual simulation effects (rain, skybox) based on real timelines.
- Traffic Anomaly Detection: Implementation of Local Outlier Factor (LOF) algorithms to detect contextual traffic incidents (e.g., congestion during peak hours).
- Air Quality Prediction: Random Forest Classifier optimized with SMOTE (Synthetic Minority Over-sampling Technique) to predict air quality alerts with high accuracy (83%).
- Incident Visualization: Automatic instantiation of 3D incident prefabs (e.g., broken down vehicles) when ML pipelines detect outliers.