Introducing KH3: A Frugal Trajectory Indexing System
François Legras, Iwao Fujino, Thierry Le Pors, Laurent Étienne · 2025
Processing ship trajectory data using Automatic Identification System (AIS) presents significant challenges. AIS generates large volumes of spatiotemporal data, making efficient indexing and pattern detection challenging. We propose an extension of the H3 geospatial indexing system by integrating kinetic data. This integration aims to capture not only the instantaneous description of movement but also segments of trajectories. The proposed Kinetic H3 (KH3) framework is designed to be computationally efficient and to minimize the representation volume. The application of this system is further investigated through the utilization of a maritime dataset drawn from real-world scenarios. Among the explored applications, our objective is to support previous research endeavors that engage in Topic Model Analysis to identify patterns within trajectories.