Classification of Inter-Urban Highway Drivers’ Resting Behavior for Advanced Driver-Assistance System Technologies using Vehicle Trajectory Data from Car Navigation Systems

Choi, Jaeheon; Lee, Kyuil; Kim, Hyunmyung; An, Sunghi; Nam, Daisik · 2020 · Crossref

DOI: 10.3390/su12155936

archive: archived pipeline: cataloged verified

Get this paper ↗ (DOI — opens at the source; we link to it, we don't host it)

Summary

This study addresses the critical issue of fatigue-related crashes on inter-urban highways, which are primarily caused by drowsy or distracted driving. The authors aim to develop reliable criteria for Advanced Driver-Assistance Systems (ADAS) to guide drivers toward rest areas, thereby improving traffic safety and transportation sustainability. Traditional research methods, such as field surveys at specific locations like toll booths or rest areas, are often expensive, biased by sample location and time, and lack comprehensive data coverage. To overcome these limitations, the researchers utilized a large-scale dataset from car navigation systems to analyze drivers’ resting behaviors in relation to travel distances, providing a more robust foundation for safety strategies. The methodology relies on vehicle trajectory data collected from car navigation systems in Korea over a four-month period in 2014. The dataset initially contained 14.6 million vehicle trips, which were filtered to include only 591,103 trajectories involving highway travel. To minimize seasonal and weekly biases, data were restricted to March, April, October, and November, specifically Tuesday through Thursday. Resting behavior was defined by vehicles stopping at officially designated rest areas for at least five minutes. The researchers analyzed the correlation between cumulative travel distance and the number of rests, employing a statistical hypothesis test combined with a random sampling method based on Monte-Carlo simulation. This approach allowed for the categorization of drivers into distinct groups based on their driving distances and resting frequencies, while balancing sample sizes across different distance intervals to reduce outlier impacts. The results demonstrate a significant positive correlation between travel distance and the likelihood of resting. For short trips under 50 km, only 0.8% of drivers stopped at rest areas. However, as distance increased, the resting rate rose sharply; approximately 43% of drivers in the 150–200 km interval rested, and over 85% of those traveling more than 350 km stopped at least once. The analysis revealed two distinct behavioral patterns: drivers traveling less than 100 km rarely rested more than once, whereas those traveling over 200 km showed a significantly higher probability of multiple rests. The Monte-Carlo simulation successfully categorized these behaviors, confirming that cumulative travel distance is a primary determinant of resting decisions. The significance of this research lies in its provision of data-driven criteria for ADAS technologies and highway operators. By establishing a clear relationship between travel distance and resting behavior, the proposed algorithm can be integrated into car navigation systems to provide personalized, timely alerts for drivers likely to experience fatigue. This offers a mid-term solution to enhance road safety by encouraging breaks before fatigue becomes critical. Furthermore, the methodology enables sustainable traffic safety operators to update guidance strategies using real-time, large-scale data, reducing reliance on biased, costly field surveys. The findings support the development of adaptive alert systems that can improve driver compliance and reduce fatigue-related accidents on inter-urban highways.

Provenance

The full processing record for this entry. Every stage of this paper's journey through the pipeline is logged — what ran, with which tool and model, how many attempts it took, and when it last completed.

Summary generated by qwen3.6-27b-prismaquant on 2026-06-10; verification: verified.

Topics

Ranked by relevance to this paper. Hover a topic for its definition.

• drowsiness detection algorithms
• exposure measurement
• truck driver fatigue

Information type

What kind of knowledge this paper contributes, grouped by family — independent of topic (what it is about) and method (how it was studied).

• Empirical Findings: observational prevalence
• Methodological Resource: dataset resource
• Theoretical Contribution: computational model