Skip to content

Real-time maritime decision support system integrating AIS data, weather information, risk prediction, and voice interaction.

Notifications You must be signed in to change notification settings

SaraDeveloperX/IMS-Project

Repository files navigation

Intelligent Maritime System (IMS)

Real-Time AIS and Weather Fusion with Transformer-Based Maritime Risk Forecasting



Maritime Map

The IMS dashboard provides real-time AIS vessel positioning combined with high-resolution weather data through a MapLibre-based interface. The system supports situational awareness by integrating dynamic vessel movements and short-range environmental conditions.


System Architecture

The architecture integrates AIS ingestion, weather retrieval, and Transformer-based risk inference into a unified maritime intelligence platform optimized for operational environments.


Data Sources

IMS relies on two primary external data providers:

AIS Data Stream

Real-time AIS messages are obtained from AISstream, which provides continuous vessel positions, movement updates, and navigational status.
Official source:
https://aisstream.io

Weather Data

High-frequency weather data, including wind speed, gusts, temperature, and precipitation, is retrieved from Open-Meteo.
Official source:
https://open-meteo.com

These two real-time data providers supply the contextual AIS–weather windows used for risk estimation and short-term hazard prediction.


Demonstration Video

A complete demonstration of the system is available here:
View the demo video


System Summary

  • AIS Integration: Real-time positional data from AISstream
  • Weather Integration: High-resolution forecasts from Open-Meteo
  • Frontend: MapLibre dashboard for AIS and weather fusion
  • Backend: REST APIs and orchestration services
  • Risk Model: Transformer-based microservice for short-term hazard prediction
  • Data Fusion: Combined AIS–weather inputs for multi-factor risk scoring

About

Real-time maritime decision support system integrating AIS data, weather information, risk prediction, and voice interaction.

Topics

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published