Recognition method of construction conflict based on driver's eye movement.

Area of Science:

• Human-Computer Interaction
• Traffic Safety Engineering
• Cognitive Psychology

Background:

• Construction zones present significant risks to drivers due to unpredictable conditions.
• Understanding driver behavior, specifically eye movements, is crucial for enhancing safety in these environments.

Purpose of the Study:

• To establish a relationship between driver eye movement patterns and the occurrence of construction conflicts.
• To develop and validate a novel method for recognizing construction conflict periods.

Main Methods:

• Collected and analyzed driver eye movement data in simulated construction zones at various speeds.
• Identified key eye movement indicators: peak wavelet-processed pupil diameter, pre-peak left-side point, and post-peak blink.
• Developed the Construction Conflict Recognition Method (CCFRM) based on these key points and the GSA algorithm.

Main Results:

• The CCFRM effectively identifies construction conflict periods using selected driver eye movement key points.
• Validation demonstrated good performance in terms of recognition speed and location accuracy.
• The study confirmed the feasibility of using these specific eye movement patterns for conflict detection.

Conclusions:

• The proposed key points derived from driver eye movements are effective for recognizing construction conflict periods.
• The developed CCFRM offers a viable and accurate approach to improving traffic safety in construction zones.
• This research provides a foundation for real-time driver behavior monitoring and intervention systems.