Abstract: Pilots need to simultaneously process multiple information sources and tasks while performing tasks, which increases the workload of mental labor. In order to improve flight safety and pilot work efficiency, a multidimensional comprehensive evaluation model for pilot mental workload based on random forest algorithm is studied. A linear finite pulse response bandpass filter is used to process EEG(Electroencephalogram) signals, removing high-frequency and low-frequency noise, calculating mismatched negative waves, obtaining linearly interpolated EEG signal sampling points, and extracting power spectral density and energy features of each rhythm based on overlapping sampling points in the EEG signal neighborhood. A multi-dimensional comprehensive evaluation model of the random forest algorithm is constructed, determine the output points of each signal fluctuation frequency, and combine the voting mode to obtain the optimal classification results of multi-dimensional mental load, achieving comprehensive evaluation of pilot mental load. The experimental results demonstrate that the proposed method has high classification accuracy and can accurately evaluate the mental workload status of pilots.

Key words: random forest algorithm, brain load, multidimensional comprehensive evaluation, power spectral density, mismatched negative wave