100-Car Naturalistic Driving Study: Risk Assessment of Unsafe Driving Behaviors

A dataset from the 100-Car Naturalistic Driving Study funded by the AAA Foundation for Traffic Safety. The data were collected by the Virginia Tech Transportation Institute under sponsorship from NHTSA, VDOT, and Virginia Tech. The primary aim was to understand the relationship between potentially unsafe driving behaviors and crashes, near-crashes, and incidents.

Use Cases

• Modeling crash risk based on specific driving states like drowsiness or inattention mentioned in the description
• Quantifying the odds ratio of crashes associated with aggressive driving behaviors described in the report
• Training classifiers to identify high-risk driving scenarios based on speed, attention, and drowsiness factors
• Benchmarking safety algorithms against real-world near-crash and incident data referenced in the study

Strengths

• Data originates from a landmark, instrumented vehicle study (the 100-Car Study) conducted by a major research institute (VTTI)
• Report provides specific, quantified risk associations (e.g., odds ratios with confidence intervals) for four key driving behaviors

Limitations

• Column-level documentation is absent; field semantics must be inferred after download
• Row count is unknown, which may limit suitability assessment
• Last update date is unknown; freshness unverified

Provenance

Source

Virginia Tech Transportation Institute (VTTI), sponsored by NHTSA, VDOT, and Virginia Tech

Collection Method

Data collected during the instrumented 100-Car Naturalistic Driving Study

Time Range

null

Freshness

null

Geography

null