Recognising situation awareness of drillers based on eye-tracking and EEG features

Situation awareness is crucial for the safety of oil and gas drilling operations, as deficits can increase human error rates. Current research on multimodal physiological signal modelling for drillers remains limited, and single-modality approaches struggle to comprehensively reflect the entire cognitive process from perception to decision-making. To overcome this limitation, this study combines eye-tracking metrics reflecting external gaze behaviour with electroencephalography features reflecting internal neural activity for recognising situation awareness in drillers. Through a simulated drilling monitoring experiment and machine learning model construction, results revealed significant differences in multiple eye-tracking and electroencephalography features across different situation awareness levels. The multimodal fusion model achieved a recognition accuracy of 89.33%, outperforming single-modality models. This study systematically validates the effectiveness of eye-tracking-electroencephalography fusion for situation awareness recognition among drillers. It provides scientific support for personnel training in the drilling industry and demonstrates its potential for enhancing safety in high-risk industrial environments.