Research on Pilot Angle Adaptive Decision Making
DOI:
https://doi.org/10.61173/zk1tyr10Keywords:
Computer vision, natural language process, reinforcement learningAbstract
This paper presents a comprehensive analysis of the decision-making process for pilots in modern aviation. Various studies based on industry-wide statistics have demonstrated that in the past 15 years, more than 65% of aviation accidents are caused by poor human judgment. This issue is further compounded by the increasing trend of single-pilot operation in the aviation industry. This study is focused on the cognitive model of decision-making in aviation and compares three main types of cognitive models, which are experience-based, rule-based and more adaptive scenario-based decision-making. The results show that scenario-driven strategies, which require the assessment of the real situation at the moment and mental simulation, are more effective in dealing with novel timecritical emergencies. However, this method is significantly underused. Our research shows that it is applied in only 10% of the cases. Therefore, this paper analyzes the practical application of Naturalistic Decision-Making (NDM) theory. Firstly, the theoretical basis of the NDM model is discussed. Secondly, the influencing factors of the judgment accuracy of personnel are studied. These factors are the characteristics of personnel, the dynamic environment and the design of technical systems. Finally, the practical effect of various kinds of decision-support system in different periods is discussed.