Review and Perspective for Distance-Based Clustering of Vehicle Trajectories

Abstract

In this paper, we tackle the issue of clustering trajectories of geolocalized observations based on the distance between trajectories. We first provide a comprehensive review of the different distances used in the literature to compare trajectories. Then, based on the limitations of these methods, we introduce a new distance: symmetrized segment-path distance (SSPD). We compare this new distance to the others according to their corresponding clustering results obtained using both the hierarchical clustering and affinity propagation methods. We finally present a python package: trajectory distance, which contains the methods for calculating the SSPD distance, and the other distances reviewed in this paper.