Unified Multimodal Vessel Trajectory Prediction with Explainable Navigation Intention

TL;DR

This paper introduces a unified multimodal vessel trajectory prediction framework that incorporates explainable navigation intentions, improving prediction accuracy and interpretability across diverse maritime scenarios.

Contribution

The paper presents a novel framework combining sustained and transient navigation intentions with advanced modeling techniques for enhanced vessel trajectory prediction and explainability.

Findings

Achieves significant improvements in ADE and FDE metrics.

Demonstrates broad applicability across diverse maritime scenarios.

Enhances explainability by revealing underlying navigation intentions.

Abstract

Vessel trajectory prediction is fundamental to intelligent maritime systems. Within this domain, short-term prediction of rapid behavioral changes in complex maritime environments has established multimodal trajectory prediction (MTP) as a promising research area. However, existing vessel MTP methods suffer from limited scenario applicability and insufficient explainability. To address these challenges, we propose a unified MTP framework incorporating explainable navigation intentions, which we classify into sustained and transient categories. Our method constructs sustained intention trees from historical trajectories and models dynamic transient intentions using a Conditional Variational Autoencoder (CVAE), while using a non-local attention mechanism to maintain global scenario consistency. Experiments on real Automatic Identification System (AIS) datasets demonstrates our method's broad applicability across diverse scenarios, achieving significant improvements in both ADE and FDE. Furthermore, our method improves explainability by explicitly revealing the navigational intentions underlying each predicted trajectory.