Instantaneous mental workload level recognition by combining kernel fisher discriminant analysis and Kernel Principal Component Analysis

High risk operating task in complex human machine system is vulnerable to the operator's break-down of functional state, adaptive automation system can avoid this problem. The essence of adaptive automation system is to well classify the mental workload based on electrophysiological signals. But high dimension of Electrophysiological signals made the problem difficulty. In this paper, the KFDA and KPCA is adopted to classify the OFS data from independently designed and completed experiments, and higher classification accuracy results are obtained.