🔧 Key Features

Key

• Have
• In progress / Future / Nice to have

Python FastAPI Service

• Loads trained PPO models
• RESTful action prediction
• Health monitoring
• Comprehensive error handling
• Interactive API documentation (Swagger UI)
• Configurable via environment variables
• Request/response caching for identical observations
• Model versioning and switching at runtime
• Metrics collection (request counts, latency)
• Structured logging with log levels

Java API Gateway

• REST API endpoints for traffic control
• Communicates with Python service
• Health checks and monitoring
• Robust error handling
• Service dependency management
• Load testing capabilities
• Circuit breaker pattern (Resilience4j)
• Load testing capabilities
• API request rate limiting
• Request/response logging and audit trail
• Metrics export (Micrometer/Prometheus)
• Request validation and sanitization

SUMO Traffic Simulator Integration

• SUMO TraCI (Traffic Control Interface) integration
• Vehicle simulation and tracking
• Traffic light control via SUMO
• Real-time traffic state feedback
• SUMO network configuration
• Vehicle data collection and logging
• Simulation metrics and reporting
• Integration with RL inference service

Security & Authentication (Optional - A00163691-JWTAuth Branch)

• JWT token generation and validation
• Bearer token authentication for API endpoints
• Token refresh mechanism
• Role-based access control (RBAC)
• API key management
• Request signing with JWT
• Token expiration and revocation
• Secure token storage (in transit with HTTPS)
• Rate limiting per authenticated user
• Audit logging for authentication events

LSTM Traffic Predictor Service

• Load SUMO historical data (edgeData.xml)
• Preprocess time-series data (stationarity, scaling, sliding windows)
• Train LSTM model on historical traffic patterns
• Achieve MAE < 10% forecasting accuracy
• Predict vehicle flow 15 minutes ahead
• Handle missing sensor data (KNN imputation)
• REST endpoints (health, forecast, model_info)
• Interactive API documentation (Swagger UI)
• Health monitoring and logging
• Docker containerization
• Integration with RL Inference Service
• Bidirectional LSTM for enhanced accuracy (future)
• Attention mechanism for temporal weighting (future)
• Real-time model retraining capability (future)
• Weather data integration (future)

Docker Integration

• Multi-stage builds for optimization
• Service orchestration with Docker Compose
• Health checks and automatic restart
• Persistent volumes for models
• Network isolation
• Logging and monitoring
• Kubernetes deployment manifests (optional)
• Helm charts for production deployments
• Resource limits and requests configuration
• Container registry (Docker Hub/ECR)

Deployment

• Cloud deployment on Render
• GitHub Actions CI/CD pipeline (optional)
• Environment-based configuration (local/production)
• Graceful service fallback and degradation
• Model hot-loading without service restart
• Rollback mechanism for failed deployments
• Automated database migrations (if applicable)
• Secrets management (environment variables/vaults)
• Performance monitoring and alerting

Monitoring & Observability

• Prometheus metrics export
• Grafana dashboards for visualization
• Distributed tracing (Jaeger/Zipkin)
• Centralized logging (ELK Stack/Loki)
• Alert rules and notifications
• SLA/SLO tracking
• Custom business metrics

Documentation

• MkDocs with Material theme
• Auto-generated API documentation (Swagger/OpenAPI)
• Complete setup and troubleshooting guides
• Architecture and design documentation
• Quick start guides (5-minute setup)
• GitHub Pages hosting
• Inline code documentation and docstrings

GitHub Pages Site

• MkDocs Material theme
• Auto-generated from main branch docs
• Published at: https://joeaoregan.github.io/TUS-26-ETP-AI-Traffic-Optimisation/
• Search functionality
• Versioned documentation (multiple versions)
• Custom domain setup (optional)