| Human factors in aviation is an important issue for improving aviation safety.Pilot's workload research is one of the most important research areas of human factors in aviation.Research on pilot's workload can improve the design of cockpit man-machine interface and reduce the proportion of man-made aviation accidents.Based on the crew functions of the smallest flight crew in China Civil Aviation Regulations,the pilot's workload evaluation system is proposed under the crew function distribution,which could provide theoretical basis and technical guidance for improving aviation safety.In this paper,the typical flight tasks under normal conditions and emergency situations are selected.The subjects are all experienced flight crews,and the experimental datasets and feature extraction sets of typical flight tasks are established.The energy features based on EEMD wavelet singular entropy and the general feature-based statistical extraction method are proposed.The proposed method is suitable for the real pilot's multiple physiological signals parameters and flight data process.The result of the flight performance example has demonstrated that the proposed feature extraction algorithm proves its effectiveness and superiority.Then,this paper establishes a crew functional classification model based on Wavelet-SVM with PSO algorithm optimization.Compared with the distribution of crew functions under real conditions,the experimental results have verified the feasibility of the proposed crew functional classification model.At the same time,the experimental results also give the optimal combination of features under normal conditions and emergency situations.Finally,under the distribution of different crew functions,the paper analyzes the correlation between effort pilot's workload and time pressure pilot's workload.Under this case,it has established the pilot's workload evaluation system,which contributes to establishment of crew functional classification based on big data in aviation. |
PDF Full Text Request