Marine fishing boat trajectory prediction method and system based on Beidou and AIS data fusion

Abstract

The invention relates to the field of marine fishery informatization and discloses a marine fishing vessel trajectory prediction method and system based on Beidou and AIS data fusion so as to improvemarine fishery vessel monitoring and an anomaly detection level and ensure safe operation of a fishing vessel. The trajectory prediction method comprises steps of inputting observation value time sequences of Beidou and AIS, and synchronizing observation data of each sensor for the same target to the same timeline; fusing the AIS data and the Beidou data according to the result of time registration of AIS trajectory data and Beidou trajectory data; according to the result obtained after the AIS data and the Beidou data are fused, obtaining optimal state estimation at the current moment, and performing trajectory prediction on the basis of the optimal state estimation. The method is advantaged in that the Beidou and AIS data are fused, fishing boat track prediction is realized based on theBeidou and AIS data, prediction precision is high, applicability is high when data transmission is lost, and a marine fishery ship monitoring level is effectively improved.

Description

technical field [0001] The invention relates to the technical field of marine fishery informatization, and more particularly to a method and system for predicting the trajectory of a marine fishing vessel combining Beidou and AIS. Background technique [0002] In the field of marine fishery safety monitoring, it is of great significance in ship state monitoring and anomaly detection to predict the state of the ship at the next moment through the historical ship trajectory data and obtain the ship trajectory prediction result. [0003] Fishing vessel trajectory prediction can not only estimate the state of the fishing vessel when the monitoring data transmission of the fishing vessel is lost, but also detect the possible abnormal behavior of the fishing vessel in advance, monitor the fishing vessel, and warn the abnormal navigation behavior. Traditional trajectory prediction methods, such as trajectory prediction based on Gaussian mixture model, moving object position predict...

AI Technical Summary

Problems solved by technology

Traditional trajectory prediction methods such as trajectory prediction based on Gaussian mixture model, location prediction of moving objects based on frequent trajectories, trajectory location prediction based on neural network, and trajectory prediction based on Kalman filter cannot effectively realize the trajectory prediction of fishing boats.

The first three trajectory prediction methods are based on the fixed trajectory behavior pattern in a specific area. For example, the trajectory on land follows the road network traffic, and the trajectory on the sea is restricted by the waterway, while the fishing boat is not restricted by the waterway when sailing at sea, and its navigation behavior has great influence. In most cases, fishing boats move nonlinearly at sea, and it is impossible to extract relevant behavior patterns from historical trajectories to predict fishing boat trajectories.

The trajectory prediction method based on Kalman filter is a trajectory prediction method based on target motion characteristics and historical trajectory information, but it is only suitable for linear systems and cannot be directly applied to nonlinear systems like fishing boats

[0004] Both Beidou and AIS systems can monitor the position of fishing boats and obtain a series of time-space continuous fishing boat trajectory data, but the sensors used in the above systems will have their own measurement errors when monitoring the status information of fishing boats such as position, speed, and direction.

Images