Health in Emergencies & Disasters Quarterly (Jul 2024)

Investigating the Relationship Between Subjective and Interpretive drowsiness With Lane Departure in Simulator Driving

  • Ali Askari,
  • Robab Hosseinpour,
  • Mohammad Bakhtiari,
  • Parvin Sepehr,
  • Abbas Ghodrati Torbati,
  • Ali Salehi Sahlabadi,
  • Maliheh Eshaghzadeh,
  • Anahita Zandi,
  • Javad Vatani,
  • Mohsen Poursadeghiyan

Journal volume & issue
Vol. 9, no. 4
pp. 255 – 264

Abstract

Read online

Background: Driver drowsiness is a significant factor contributing to road accidents and the overall increase in road mortality rates. This study investigates the relationship between lane crossing (LC), the number of lateral position deviations and driver drowsiness. The proposed method, due to precision and convenience, has great potential for developing driver-assistant systems. Materials and Methods: In this experimental research, 34 sub-urban bus drivers participated in a 2-h driving session in a simulator designed based on visual reality. Sensors attached to the steering recorded right and left deviations and relevant information was matched with the receded videos and the amount of the standard deviation of lane position (SDLP). The number of LC was determined with the designed indicator on the road software. Then, the association between SDLP and the number of LC was compared with the results of the Karolinska sleepiness scale (KSS) and observer rating of drowsiness (ORD), which determined the level of drowsiness, by facial features. Results: The results of multivariate analysis of variance indicated that the time variable has a significant effect on both ORD and SDLP (P<0.05). These two variables provided over 99% of the variance. The same results were obtained for KSS and SDLP (P<0.05). Meanwhile, the linear combination of these two dependent variables over 12 periods of the research has significant variations. In addition, the results show the progression of KSS and ORD (P<0.05) and an increase in SDLP and LC (P<0.05). In the same manner, LC has a tight association with the level of drowsiness and other factors (P<0.05). Conclusion: Drowsiness increases the variation in line tracking. However, it is not an appropriate signal for drowsiness detection. The SDLP and the number of line crossings is an appropriate criterion to check drivers’ performance.

Keywords