Mobile-based wearable-type of driver fatigue detection by GSR and EMG

Abstract

Driver fatigue had been a major reason that leads to road accidents. This paper focuses on investigating the usage of electromyography and galvanic skin response to detect the driver fatigue symptoms. The study reveals that the variation of EMG signal patterns can be mapped to simulate the driving behavior. In the study, five attachment positions of EMG sensors are observed to indicate the best position of the mapping for wheel steering control behavior. Hereby, the study also reveals that the changing variations of EMGs in frequency-domain are excellent and significant fatigue indicator than the usual time-domain features. On the other hand, existing systems only focused on analyzing the signal pattern of GSR, but not the variation of GSR in accordance to frequency analysis, which is one of our main objectives study. The sensed EMGs and GSRs are transmitted to the mobile device via Bluetooth Low Energy. The analysis takes part in mobile device with implemented fatigue monitoring application. If the developed classifier indicates the driver vigilance level dropped to dangerous predefined threshold, a vibration warning will be triggered to alert the driver. In fact, the experiment results revealed that the significant differences in EMG and GSR features are managed to determine the driver fatigue in five seconds interval. The developed SVM classifier of mobile application shows average of 92% fatigue detection accuracy rate.

Cite this work

No public copy is stored for this record, so only the metadata is cited.

Stored only when both papers are in the corpus — 59 citations exist upstream.

Papers that cite the same references — the relational fallback while embeddings are still backfilling.