Residual MLP Network for Mental Fatigue Classification in Mining Workers from Brain Data

摘要: At the mining industry, human safety and productivity are both desirable in logistics pipeline. Since operation of heavy machines requires continued vigilance mental activity, fatigue caused by long hours work constant effort generally occurs this environment. In general, is related to a loss efficiency, leading decrease inducing critical errors, which can provoke equipment breakups or accidents with victims. high cognitive workload environment, there need for development robust monitoring techniques aiming predict before workers' movement responses become slower, more variable, error-prone. work, we introduce residual multilayer perceptron (MLP) network (ResMLPNet) assess its performance challenging problem classification from electrophysiology data, acquired during Virtual Reality (VR) training sessions mimicking real faced excavator workers at industry. three-step strategy, ResMLPNet achieved slightly better accuracies compared plain MLP architecture.