Trajectory Prediction Based on Machine Learning
Lumin Su, Li Li · IOP Conference Series Materials Science and Engineering · 2020
Abstract With the rapid development of the Internet and positioning technology, trajectory prediction has become a popular research direction because of its wide range of application scenarios. However, the trajectory points are represented only by a series of incomprehensible numerical labels in current studies, and the extraction of the stay points are also lack of time consistency. On the other hand, conventional trajectory prediction models do not make full use of the trajectory context information, as well as only predict the position of the next stay points, which limits the prediction accuracy. To fulfill the above problems, the trajectory prediction method based on machine learning is studied. In this paper, a spatial information extraction method based on multistage cluster is proposed, as well as the LSTM model and the bidirectional LSTM model in the deep learning are used to predict the next position. Experimental results show that the novel method proposed can better represent the semantic information of trajectory and improve the prediction performance greatly.