BATMAN: A Brain-like Approach for Tracking Maritime Activity and Nuance

Area of Science:

• Geospatial Intelligence
• Artificial Intelligence
• Maritime Security

Background:

• Increasing maritime traffic volume necessitates advanced analysis methods.
• Anomalous maritime events require identification for law enforcement, government, and military interests.
• Commercial geospatial intelligence data availability is rising.

Purpose of the Study:

• To develop a data fusion pipeline for identifying ships and classifying their behaviors at sea.
• To integrate artificial intelligence with traditional algorithms for enhanced maritime surveillance.
• To reduce human workload in analyzing maritime activities.

Main Methods:

• A data fusion pipeline combining visual spectrum satellite imagery and Automatic Identification System (AIS) data for ship identification.
• Integration of contextual environmental data (e.g., EEZ boundaries, undersea cables, weather) for behavior classification.
• Utilizing freely accessible data sources like Google Earth and US Coast Guard data.

Main Results:

• Successfully identified ships at sea by fusing satellite imagery and AIS data.
• Classified various maritime behaviors including illegal fishing, trans-shipment, and spoofing.
• Demonstrated the pipeline's capability to use contextual information for nuanced behavior analysis.

Conclusions:

• The proposed pipeline is the first to move beyond basic ship identification to tangible behavior classification.
• This AI-driven approach aids analysts in identifying suspicious activities and reduces manual effort.
• The framework offers a cost-effective solution for maritime security and monitoring.