Real-World Marine Radar Datasets for Evaluating Target Tracking Methods.

Area of Science:

• Maritime autonomous systems
• Robotics and intelligent systems
• Signal processing and data analysis

Background:

• Increasing implementation of autonomous navigation in maritime domains necessitates robust tracking for enhanced situational awareness.
• Accurate target detection and tracking are crucial for maritime traffic monitoring and safety.

Purpose of the Study:

• To introduce a novel online repository of marine radar datasets for target detection and tracking research.
• To present and evaluate a new extended centroid-based multiple target tracking algorithm.

Main Methods:

• Collected and curated three marine radar datasets from real-world measurement campaigns.
• Utilized Automatic Identification System (AIS) for reference positions.
• Developed a novel extended centroid-based multiple target tracking algorithm and compared it to its standard version.

Main Results:

• The novel algorithm demonstrated performance improvements over the standard version on the provided datasets.
• Initial dataset-specific analysis provided insights into algorithm performance under various conditions.

Conclusions:

• The developed datasets and algorithm contribute to advancing research in maritime autonomous navigation and traffic monitoring.
• The presented resources and methods offer a valuable tool for researchers in target detection and tracking.