Options
2014
Conference Paper
Title
Semi-automatic sea lane extraction combining Particle Filtering (PF) and Geographic Information Systems (GIS)
Abstract
This paper addresses the problem of extracting traffic patterns from noisy and incomplete sensor data. We introduce an approach that combines the output of (particle filter based) tracking filters with GIS techniques. While tracking filters are standard tools for generating estimates of the position and velocities of individual objects over time, GIS techniques are used to extract geographic patterns from these estimates. GIS techniques also allow the integration of prior knowledge, such as coastlines and bathymetry, to enhance the quality of the extracted patterns. The approach has been applied to maritime traffic surveillance using report of the Automatic Identification System (AIS), which may be subject to GPS errors, spoofing, time delays and periods of missed detections. The results are continuous vessel tracks indicating each vessel's route, even when the update rate of messages is significantly slower.
Conference