Search for a command to run...
This study proposes a high-efficiency framework for identifying ship encounter scenarios using compressed Automatic Identification System (AIS) trajectories, addressing the computational and scalability limitations of existing methods that rely on complete trajectories and high-frequency interpolation. By combining Douglas-Peucker (DP)-based trajectory compression with bounding-box filtering and lightweight post-compression interpolation, the framework exploits the sparsity of compressed trajectories to improve computational efficiency, after which encounters are identified using spatiotemporal proximity metrics. To account for information degradation introduced by trajectory compression, a multi-dimensional evaluation process integrating behavior analysis and a fuzzy comprehensive collision risk index (CRI) is developed to assess scenario reliability in terms of identification accuracy, behavioral fidelity, and risk consistency. The evaluation also incorporates Bayesian inference to characterize uncertainty from compression and scenario heterogeneity, providing a probabilistic description of compression-induced errors. A case study using AIS data from the Gulf of Finland demonstrates that the proposed framework can substantially reduce data volume and computational cost. Experiments with six mainstream DP-family compression algorithms show that advanced DP variants can maintain high scenario consistency while preserving encounter geometries and risk evolution patterns. These findings highlight the practicality and reliability of the framework for efficient encounter identification in large-scale maritime traffic analysis. • Proposes a framework for encounter identification on compressed trajectories. • Uses bounding boxes and lightweight interpolation to exploit trajectory sparsity. • Develops a Bayesian reliability evaluation process for the identified scenarios. • Compares six DP-family algorithms to assess practical performance and reliability.