Leveraging AIS data for maneuver detection and knowledge extraction during ship encounter

• Identifying maneuvers made to avoid risk or ensure safe passage during ship encounters is essential for understanding ship behavior patterns. These maneuvers reflect the underlying decision-making processes and can be analyzed to enhance our understanding of ship behavior in encounter situations. • A novel method is proposed to extract encounter-related maneuvers by detecting ship motion alterations and applying logic rules for maneuver identification. • Knowledge extraction and presentation through statistical analysis of the extracted decision-making data, based on AIS data from Oslofjord, Norway, demonstrates the method’s potential to advance understanding of ship decision-making and support improved situational awareness. Understanding the ship encounter process has become increasingly important for improving maritime situation awareness and intelligent navigation. Maneuvers made during the decision-making process play a crucial role in understanding how ships make decisions for safe operations. This necessitates the detection of encounter-based navigation maneuvers from ship trajectory data, which is challenging due to the complexity of the data and the relationships between encountering ships. Thus, this paper proposes a novel method to detect ship maneuvers during encounter situations from historical AIS (Automatic Identification System) data. The method detects ship maneuvers by identifying motion alterations while mitigating confusion caused by motion changes due to other external factors. Additionally, logic rules are introduced to further evaluate detected motion alterations that qualify as encounter-related maneuvers. Following the maneuver detection, statistical analysis is conducted based on the detected results for knowledge extraction. The analysis focuses on two key aspects—ship decision-making and ship encounter processes—to extract insights to enhance the understanding of the ship decision-making process to keep safety. The AIS data, collected from Oslofjord, Norway, demonstrates the effectiveness of the proposed method. A case of knowledge application illustrates the potential to aid in improving maritime situation awareness and decision-making process.