Ship trajectory prediction method based on bayesian optimization of directional inertia correction

By employing a Bayesian-optimized orientation inertia correction method, combined with inertia decay, orientation calibration, and affine coordinate transformation, the problem of insufficient accuracy of traditional models in complex marine environments is solved, achieving efficient and interpretable ship trajectory prediction, which is suitable for maritime traffic management.

CN122410992APending Publication Date: 2026-07-17JIMEI UNIV
View PDF 0 Cites 0 Cited by

Patent Information

Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
JIMEI UNIV
Filing Date
2026-04-21
Publication Date
2026-07-17

AI Technical Summary

Technical Problem

Existing ship trajectory prediction technologies struggle to achieve an ideal engineering balance between accuracy, efficiency, and interpretability. Traditional models cannot adapt to the dynamic nature of complex marine environments, while deep learning models suffer from high computational costs and poor interpretability, failing to meet the needs of maritime traffic safety early warning and management.

Method used

The Bayesian optimization-based orientation inertia correction method for ship trajectory prediction achieves high-precision prediction with a lightweight model by automatically determining parameters through inertia decay, orientation calibration, affine coordinate system transformation, and trajectory smoothing mechanisms, combined with the Bayesian optimization algorithm.

Benefits of technology

It achieves high-precision, low-computational-cost ship trajectory prediction in complex marine environments, making it suitable for deployment on shipborne edge devices with limited computing resources, and providing reliable maritime traffic safety support.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN122410992A_ABST
    Figure CN122410992A_ABST
Patent Text Reader

Abstract

The application provides a ship trajectory prediction method based on Bayesian optimization of directional inertia correction, comprising: acquiring a historical trajectory sequence of a ship containing time-sequentially arranged latitude and longitude coordinate points; calculating an instantaneous speed vector containing a latitude component and a longitude component based on the historical trajectory sequence of the ship; multiplying the latitude component of the instantaneous speed vector by the i-th power of an inertia attenuation factor to obtain a latitude direction correction speed for the i-th prediction step; multiplying the longitude component of the instantaneous speed vector by the product of the i-th power of the inertia attenuation factor and a directional inertia factor to obtain a longitude direction correction speed; wherein the inertia attenuation factor represents the exponential attenuation characteristic of the inertia of the ship movement with the prediction step, and the directional inertia factor represents the scale anisotropy of the longitude direction and the latitude direction in the earth spherical coordinate system; and sequentially calculating the predicted positions of a plurality of future prediction steps based on the latitude direction correction speed and the longitude direction correction speed to obtain a future trajectory sequence of the ship.
Need to check novelty before this filing date? Find Prior Art