This study proposes a new adaptive ship domain model based on the Rate of Turn (ROT) to determine the final collision avoidance capability of a ship. Existing studies on ship domains have primarily focused on static factors such as ship length and spe...
This study proposes a new adaptive ship domain model based on the Rate of Turn (ROT) to determine the final collision avoidance capability of a ship. Existing studies on ship domains have primarily focused on static factors such as ship length and speed, lacking direct incorporation of ROT,which is the most crucial indicator of a ships maneuverability in actual collision avoidance situations. Therefore, this study introduces ROT as a key dynamic variable to define a safety domain that reflects the vessels actual physical limits. The research methodology involves designing 159 precise encounter scenarios based on COLREGs (Rules 14 and 15). A kinematic simulation engine was developed to model the vessels trajectory. This model assumes the own ship performs a 60° starboard avoidance maneuver based on its- vii 30°/min ROT capability, while the target ship remains non-cooperative. A binary search algorithm is applied to this engine to progressively find the last-point to avoid collision for each scenario, thereby defining the minimum required physical boundary. The results demonstrate that the developed ROT-based domain is fundamentally different from traditional static models. It exhibits significant anisotropy (elongated forward) and, critically, asymmetry (wider on the port side). This asymmetry is a direct consequence of the COLREGs-mandated starboard-turn rule, proving that the risk is not
symmetrical as assumed in previous models. The findings of this study are expected to contribute to a more accurate determination of collision avoidance possibility. This adaptive domain, which is based on the ships actual maneuvering performance rather than statistical averages, can serve as a core algorithm for autonomous ship navigation systems (MASS) and provide a more realistic decision-support tool for human navigators.