Classification of Mental Load of Special Vehicle Crew Based on Convolutional Neural Network

Detecting the mental load state of special vehicle crew is of great significance to monitor the driving state of crew and improve the comprehensive combat effectiveness of crew. Based on the virtual simulation system of crew task of an special vehicle, 12 subjects were selected to carry out the brain load experiment of special vehicle commander for the new task of special vehicle commander. The experimental results show that adding sub tasks to typical combat tasks has higher scores on Subjective scales and lower accuracy of typical combat tasks than only completing typical combat tasks; Compared with the attack and report stage, the search stage has higher scores of Subjective scales and absolute power of theta band, alpha band and beta band. A feasible model of deep learning network based on EEG to detect the mental load of special vehicle passengers is established. The deep learning network model is based on THE RESNET convolutional neural network. The results show that the deep learning model can effectively extract the characteristics of EEG and realize the classification of crew mental load.