A weighted method modelling the trajectory uncertainty based on moving objects’ speeds and positions

Trajectory data inevitably contain various errors which would affect the effectiveness of the trajectory data in various applications. Therefore, trajectory uncertainty modelling is a prerequisite problem that needs to be solved. Trajectory uncertainty is generally estimated by using only speeds or positions of the moving objects. However, both of the speeds and positions of the moving objects have influences on the trajectory uncertainty. Therefore, in this paper, a speed and position weighting (SPW) model is designed to integrate the classical speed based model and the geometric distance based model by using an adaptive weighting strategy. The weights of these two models are determined based on the relationship between the Euclidean distance and the distance of the uniform motion between sampled points. Two indicators of uncertainty area proportion and the point coverage are adopted to validate the effectiveness of the SPW model. The results of the experiment prove that the SPW model can provide a moderate uncertain region contain almost the whole trajectory.