Modeling Human Encounter Situation Awareness Results Using Support Vector Machine Models

Abstract

This study constructs a support vector machine model based on supervised learning to model the results of situation awareness for ship collision avoidance. To explain the model, collision risk situations were defined, and human situation recognition results were collected in the specified cases. Moreover, it was used to build predictors and outcome variables. Finally, the constructed variable was applied to the classification model. This model provides insight into the results of the navigator’s encounter situation awareness when collision avoidance is required. The results indicate that the proposed model can be used to predict human situation awareness outcomes in given cases.