An Intelligent Maritime Anomaly Detection Approach Using AIS and Environmental Data

Abstract

With the rapid growth of global maritime transportation, ensuring navigational safety has become increasingly critical. Traditional vessel monitoring methods primarily rely on single-source data such as AIS, which are insufficient for capturing complex maritime conditions. This study proposes an intelligent maritime anomaly detection approach by integrating AIS data with environmental information, including wind, wave, and sea current conditions. The proposed method employs feature engineering techniques to extract kinematic, spatial-temporal, and environmental features, followed by machine learning models such as Random Forest and Support Vector Machine for anomaly classification. A simulation-based dataset is constructed to evaluate system performance under various maritime scenarios. Experimental results demonstrate that incorporating environmental data significantly improves anomaly detection accuracy and robustness compared to AIS-only approaches. The proposed framework provides an effective and practical solution for enhancing maritime situational awareness and safety monitoring. The proposed framework demonstrates strong potential for real-world maritime surveillance and safety applications.