Dynamic Prediction Model of Situation Awareness in Flight Simulation

Abstract

Dynamic prediction for pilot situation awareness (SA) is an important issue in aviation safety. This paper presents a dynamic prediction model on the basis of the progressive triggering relationship between low and high-level SA. Six typical cognitive status (“Unnoticed”, “Attention of situation element (SE) but not reaching perception”, “Perception of SE”, “Perception but not matching the best rule”, “Triggering of the best rule” and “Decision making and operation”) were proposed for the description of the cognitive process of SE. Eighteen participants were selected to conduct the flight simulation tasks, and the situation awareness global assessment technique (SAGAT) method was adopted to measure the performance data (including accuracy and response time) at 13 typical time points. Statistical analysis showed that the theoretical value of the proposed SA dynamic prediction model was significantly correlated with accuracy and response time, which validated the model in the flight simulation environment preliminary. The proposed SA dynamic model in flight scenarios can give some references for cockpit’s human-computer interface design and flight tasks optimization assignment.