K-nearest neighbor algorithm-based meeting ship collision danger category determination method and device

The invention belongs to the field of intelligent ship navigation control, particularly relates to a K-nearest neighbor algorithm-based meeting ship collision danger category determination method anddevice, and aims to solve the problems that an existing method is complex in calculation process, large in data statistics when indexes are excessive and difficult in weight determination. The methodincludes: based on pre-acquired position information and motion information of a ship and a target ship, obtaining feature values of the plurality of meeting ship collision risk evaluation indexes, taking the feature values of the plurality of meeting ship collision risk evaluation indexes as to-be-classified data samples, and classifying the to-be-classified data samples through a pre-establishedcollision risk category classification model to obtain corresponding meeting ship collision risk categories; wherein the collision danger category classification model is established based on a K-nearest neighbor algorithm. Th