不同助航灯光情境下飞行员认知负荷识别研究  

Research on Recognition of Pilots’ Mental Workload under Different Navigational Lighting Scenarios

作　　者：焦帅涛 朱文兵 刘闯 张帅 张敬坤 张宇 张晨阳 

机构地区：[1]中国民用航空飞行学院新津分院,四川 成都 [2]西南交通大学交通运输与物流学院,四川 成都 [3]西部航空有限责任公司飞行部,重庆

出　　处：《国际航空航天科学》2024年第2期88-95,共8页Journal of Aerospace Science and Technology

摘　　要：该研究探讨了不同助航灯光情境下飞行员认知负荷的识别方法。助航灯光是飞机在夜间和复杂气象进场着陆中的重要视觉辅助工具,其失效会增加飞行员的认知负荷,影响飞行安全。本研究采用模拟机实验,设计了助航灯光完好和失效两种情况下的夜航进近着陆任务,使用便携式心电设备和NASA-TLX量表,采集了被试飞行员的PPG信号和主观负荷评分,分析了助航灯光失效对飞行员认知负荷的影响。通过提取PPG信号的HRV特征,并对比多种机器学习算法,构建了对飞行员认知负荷能有效分类的模型。结果表明,助航灯光失效显著增加了飞行员的主观认知负荷。KNN模型在识别飞行员认知负荷方面表现出最高的准确性,达70.23%。本研究的结果强调了助航灯光对保障飞行安全的重要性,研究为飞行安全管理提供了重要数据和有效工具。This study investigates methods for identifying pilots’ mental workload under different navigational lighting scenarios. Navigational lights serve as crucial visual aids for aircraft landing during night flights and complex weather conditions, with their failure increasing pilots’ mental workload and impacting flight safety. Through simulator experiments, this research designed tasks for night approach landings with both operational and failed navigational lights, collecting pilots’ PPG signals and subjective workload scores using portable electrocardiogram devices and the NASA-TLX scale to analyze the impact of navigational light failure on pilots’ mental workload. By extracting HRV features from the PPG signals and comparing various machine learning algorithms, an effective model for classifying pilots’ mental workload was constructed. The results demonstrated that the failure of navigational lights significantly increased pilots’ subjective mental workload. The KNN model exhibited the highest accuracy in identifying pilots’ mental workload, reaching 70.23%. The findings underscore the importance of navigational lights in ensuring flight safety, providing critical data and effective tools for flight safety management.

关 键 词：飞行安全 认知负荷 助航灯光 机器学习 

分 类 号：R85[医药卫生—航空、航天与航海医学]