MoCo-AIS: A Contrastive Learning Framework for Similarity Computation of Vessel Trajectories

The article introduces MoCo-AIS, a unified contrastive learning framework designed for computing similarity in vessel trajectories, leveraging the Momentum Contrast (MoCo) approach. It evaluates various deep learning models on extensive AIS datasets, demonstrating significant improvements in similarity learning performance compared to existing methods. This framework provides a standardized platform for benchmarking trajectory representation models, which is crucial for applications like mobility prediction and anomaly detection in maritime contexts.