The 100-Car Naturalistic Driving Study, Phase II - Results of the 100-Car Field Experiment

Dingus, Thomas A.; Klauer, S. G.; Neale, Vicki L. (Vicki Lewis); Petersen, A.; Lee, S. E.; Sudweeks, J; Perez, M. A.; Hankey, J.; Ramsey, D.; Gupta, S.; Bucher, C.; Doerzaph, Zachary R. (Zachary Richard); Jermeland, J.; Knipling, R. R. (Ronald R.) · 2006 · ROSA P / United States. Department of Transportation. National Highway Traffic Safety Administration

archive: archived pipeline: cataloged verified

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

Summary

The 100-Car Naturalistic Driving Study, Phase II, addresses the critical need for large-scale, real-world data on driver behavior and vehicle performance to inform intelligent vehicle safety systems. Sponsored by the National Highway Traffic Safety Administration (NHTSA) and conducted by the Virginia Tech Transportation Institute, this study was the first of its kind to collect naturalistic driving data without experimental interference. The primary motivation was to overcome the limitations of laboratory simulations and crash databases by capturing unobtrusive, longitudinal data on actual driving behaviors, including extreme cases of drowsiness, impairment, and risk-taking. The study aimed to characterize crashes, near-crashes, and incidents, quantify causal factors, and evaluate the feasibility of scaling such data collection efforts. The methodology involved instrumenting 100 vehicles, 78 of which were privately owned, with a sophisticated Data Acquisition System (DAS). The DAS included five channels of video (capturing the road, driver’s face, hands, feet, and rear view), GPS, accelerometers, Doppler radar for headway detection, and sensors for vehicle state and kinematics. Data was collected from 241 drivers over 12 to 13 months, resulting in approximately 2 million vehicle miles and 43,000 hours of driving. Drivers received no special instructions, and no experimenters were present, ensuring naturalistic conditions. The raw data was processed into an event database classifying crashes, near-crashes, and incidents based on pre-event maneuvers, precipitating factors, and avoidance maneuvers. The analysis focused on nine specific goals, including the characterization of rear-end and lane-change conflicts, the role of driver inattention, and the validation of hardware performance. Key findings revealed that driver inattention was a significant contributing factor in a substantial proportion of crashes and near-crashes. The study successfully operationalized the definition of a "near-crash" using quantitative measures such as time-to-collision and vehicle headway. Analyses of rear-end conflicts identified specific dynamic conditions and causal factors, such as following distance and lead-vehicle braking, while lane-change events were characterized by specific maneuver types and age-related risk patterns. The data also highlighted the prevalence of secondary task distractions, including wireless device use, and demonstrated that driver behavior remained consistent over time, with minimal adaptation to the instrumentation. The study confirmed that the instrumentation system was capable of capturing high-fidelity data on rare but critical safety events, providing a robust dataset for analyzing human factors in driving. The significance of this study lies in its establishment of a comprehensive, naturalistic driving database that bridges the gap between epidemiological crash data and controlled experimental studies. By providing detailed insights into the causal factors of crashes and near-crashes, the findings offer critical evidence for the development of collision avoidance systems and intelligent vehicle initiatives. The study’s success in collecting large-scale, unobtrusive data validates the naturalistic driving approach as a viable method for future safety research. Furthermore, the lessons learned regarding hardware performance, data reduction, and participant compliance provide a blueprint for larger-scale field experiments, ultimately supporting the advancement of vehicle safety technologies and the understanding of driver behavior in real-world contexts.

Key finding

The study collected approximately 2,000,000 vehicle miles and nearly 43,000 hours of naturalistic driving data from 241 drivers over 12 to 13 months, creating a comprehensive event database of crashes, near-crashes, and incidents.

Methodology

naturalistic

Sample size: 241

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. Discovered via bulk_ingest_rosap on 2026-05-23 (6 acquisition events logged).

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.

• naturalistic crash near crash
• incidence prevalence
• pre crash contributing factors
• induced exposure
• sex gender
• crash typology

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: crash risk outcomes, observational prevalence
• Methodological Resource: dataset resource