Trajectory prediction based on long short-term memory network and Kalman filter using hurricanes as an example

Wanting Qin,Songyang Lao,Cong Lu,Jun Tang

doi:10.1007/s10596-021-10037-2

Wanting Qin, Songyang Lao + Show 2 more

Open Access

https://doi.org/10.1007/s10596-021-10037-2

Copy DOI

Abstract

Trajectory data can objectively reflect the moving law of moving objects. Therefore, trajectory prediction has high application value. Hurricanes often cause incalculable losses of life and property, trajectory prediction can be an effective means to mitigate damage caused by hurricanes. With the popularization and wide application of artificial intelligence technology, from the perspective of machine learning, this paper trains a trajectory prediction model through historical trajectory data based on a long short-term memory (LSTM) network. An improved LSTM (ILSTM) trajectory prediction algorithm that improves the prediction of the simple LSTM is proposed, and the Kalman filter is used to filter the prediction results of the improved LSTM algorithm, which is called LSTM-KF. Through simulation experiments of Atlantic hurricane data from 1851 to 2016, compared to other LSTM and ILSTM algorithms, it is found that the LSTM-KF trajectory prediction algorithm has the lowest prediction error and the best prediction effect.

Highlights

With the continuous development of satellite navigation, wireless communication and other technologies, mobile intelligent devices with positioning functions are currently widely used
The trajectory map of Atlantic hurricanes from 1851 to 2015 is shown in Fig. 9 (East longitude is represented by positive values and west longitude is represented by negative values)
From the perspective of machine learning, using real Atlantic hurricane data, an long short-term memory (LSTM) network is applied to hurricane trajectory prediction, and the prediction model is trained using historical hurricane data

Summary

Introduction

With the continuous development of satellite navigation, wireless communication and other technologies, mobile intelligent devices with positioning functions are currently widely used. When people use these devices, they actively or passively record a large number of historical trajectories, leading to the formation of spatiotemporal trajectories [1, 2]. Animal migration, transportation and meteorological clouds are examples of moving objects in specific application fields. The moving object studied in this paper is the hurricane. Trajectory prediction, as the most important method to reduce the damage caused by a hurricane, has become a hot issue in the field of trajectory research.

Methods

Results

Conclusion