Constructing Maps by Clustering Trajectories

We propose a new approach for constructing the underlying map from trajectory data. This algorithm is based on the idea that road segments can be identified as stable subtrajectory clusters in the data. For this, we consider how subtrajectory clusters evolve for varying distance values, and choose stable values for these, in this way avoiding a global proximity parameter. Within trajectory clusters, we choose representatives, which are combined to form the map. We experimentally evaluate our algorithm on hiking and vehicle tracking data. These experiments demonstrate that our approach can naturally deal with different road widths, and differences in density of the data. It can also to an extent separate roads that run close to each other and can deal with outliers in the data, two issues that are notoriously difficult in road network reconstruction. Constructing Maps by Clustering Trajectories iii