A High-Precision Method for Evaluating the Similarity of Maritime Vessel Trajectories

This study investigates the demand for high-precision trajectory similarity assessment in intelligent maritime navigation. This is done by analyzing discrepancies between GPS-derived trajectories and actual vessel paths, while identifying critical limitations in existing evaluation methods. To address these challenges, we propose a robust framework that integrates three core innovations: firstly, a linear feature accuracy-constrained resampling method to ensure computational precision under diverse complexity conditions, validated through experimental verification; secondly, a shape feature extraction and transformation protocol designed to maintain consistency across multi-scale and heterogeneous operational scenarios; thirdly, a quantitative similarity evaluation criterion based on extracted shape characteristics, enabling systematic alignment between localized trajectory segments and historical navigation patterns. The experimental results confirm the method’s enhanced robustness and its capability to bridge local and global trajectory comparisons, demonstrating that shape-driven quantification significantly refines similarity analysis. This approach advances intelligent maritime systems by providing a technically rigorous solution for real-time decision support and actionable insights into next-generation navigation applications.