A trajectory similarity measurement algorithm based on three-dimensional space area division

Abstract

Abstract Aiming at the problems of most trajectory similarity measurement algorithms, such as low computational efficiency, poor robustness, and inability to distinguish trajectories with opposite directions, this paper proposes a 3D Triangle Division (3TD) algorithm. Firstly, the absolute time series of the trajectory set was transformed into a relative time series according to the time conversion rules of the 3TD algorithm. Then, in the three-dimensional space coordinate system composed of three elements of longitude, latitude, and time, the trajectories were divided into several non-overlapping triangles by partitioning rules, and the area of the triangles was accumulated and the trajectory similarity was calculated. Finally, comparative experiments with the Longest Common Subsequence (LCSS) and Hausdorff distance were carried out on a randomly sampled trajectory dataset collected from the Automatic Identification System (AIS) of ships. The experimental results show that the calculation time of the 3TD algorithm is reduced by more than 90% and the accuracy of trajectory recognition in different directions in the experimental data set is 100%. At the same time, the algorithm can also maintain accurate measurement results in the face of massive data sets and data sets with partially missing trajectory points, which can better adapt to the similarity measurement of different directions.