Using DTW to measure trajectory distance in grid space

Abstract

With the development of science and technology, we have collected a lot of trajectory data. Obtaining useful pattern from these data is an important direction of the trajectory data mining. In order to find these models, a good distance measurement between trajectories of moving object may be necessary. In this paper, we propose a new method of distance measurement, the Grid-Based DTW (GDTW). The trajectory will be converted into continuous grid cell, then get the distance by Dynamic Time Warping (DTW) measurement. This method can more effectively measure the distance of two trajectory. Experiment results show the effectiveness of our proposed means.