Vehicle Safety

Resources

The Office of Vehicle Safety Research and supports U.S. DOT’s and NHTSA’s safety goals by conducting research and safety testing of motor vehicles and motor vehicle equipment. 

NHTSA’s recently published vehicle safety reports are listed chronologically below.

69 Results
Classification of Level 2 Vehicle Events Observed On Public Roads

This report summarizes the data collected while operating three vehicles equipped with SAE automation level 2 driver assistance systems. Using cameras, driver-annotated video was recorded to document the systems’ availability and noteworthy operation. Notable events were classified into three categories: events where the vehicle terminated its automation level 2 system operation and transferred full control back to the driver (Type I), driving situations where the system remained in operation but satisfied certain classification criteria (Type II), and driving situations where either the driver performed a manual override to disengage the system, or the system automatically reestablished lane position after an unintended departure had occurred (Type III).

Driver Expectations for System Control Errors, Engagement, and Crash Avoidance in Level 2 Driving Automation Systems

This project investigated how driver expectations about Level 2 ADAS systems affect driver engagement and performance. Many vehicles available to consumers offer some level of automated driving functionality, but the capabilities of these vehicles vary widely among makes and models. Drivers may have preconceived expectations about how these driver assistance systems function. This project tested 96 participants in low- and high-capability SAE Level 2 vehicles.  Participant training evaluated if driver was in agreement with the vehicles actual capabilities (high or low) or not (i.e., they were told the vehicle had highly functioning ADAS systems when it fact it only had low, or vice versa). Testing was conducted on both public roads and on the Virginia Smart Road test track.  While on the test track, participants were asked t completed non-driving tasks while driving, and mid-way through the driving session experienced a surprise event (crash imminent scenario).

FMVSS Considerations for Vehicles with Automated Driving Systems: Phase I, Volume 2

Included in this second volume are 18 Federal Motor Vehicle Safety Standards (FMVSS) research findings, including the performance requirements and test procedures, in terms of options regarding technical translations, based on potential regulatory barriers identified for compliance verification of innovative new vehicle designs that may appear in vehicles equipped with Automated Driving Systems (ADSs). This report builds on work in the Volume 1 report (Blanco et al., 2020); that report documented the framework used to evaluate the regulatory text and Office of Vehicle Safety Compliance test procedures with the goal of identifying possible options to address unnecessary/unintended regulatory barriers for the compliance verification of ADS-dedicated vehicles (ADS-DVs) that lack manually operated driving controls.

Biofidelity Report of the THOR 5th Percentile Anthropomorphic Test Device

Three prototype anthropomorphic test devices (ATD) were fabricated to evaluate the biofidelity of a newly designed THOR 5thth percentile female ATD. This report focuses on the biofidelity responses derived from testing one of these ATDs. There were a total of 23 biofidelity test conditions that included the head, neck, shoulder, thorax, abdomen, knee-thigh-hip complex, and lower extremity. Three repeated tests were conducted on a single ATD for each test condition. The biofidelity was objectively scored in accordance with NHTSA’s Biofidelity Ranking System (BioRank). Three prototype ATDs were fabricated to evaluate the biofidelity of a newly design THOR 5thth percentile female ATD. The BioRank score of each body region was scored and the overall BioRank score of the dummy (i.e., the average of all body regions) was less than 2.0, corresponding to “good” biofidelity.

Safety of the Intended Functionality of Lane- Centering and Lane- Changing Maneuvers of a Generic Level 3 Highway Chauffeur System

This report describes the findings from applying Safety of the Intended Functionality (SOTIF) concepts as described in (ISO) Publicly Available Specification 21448 to the lane-changing and lane-centering maneuvers of a generic Level 3 highway chauffeur system. This report compares the SOTIF process described in PAS 21448 with the automotive industry’s voluntary functional safety standard, ISO 26262. This report then develops a generalized Level 3 highway chauffeur system based on discussions with stakeholders and a literature search, and identifies potential vehicle-level hazards, triggering events, and SOTIF mitigation measures for the generalized system. Finally, this report presents an approach for developing candidate scenarios to evaluate triggering events and discusses current SOTIF evaluation approaches.

Functional Safety Research Considerations for Heavy Vehicles

Industry standard ISO 26262, Road Vehicles - Functional Safety, currently applies to vehicles with gross vehicle mass up to 3,500 kg (7,716 lb), which excludes trucks and buses. This study documents factors that might necessitate functional safety approaches to have different considerations between different weight classes of vehicles and to explore how the heavy-vehicle industry is currently applying functional safety to its electrical and electronic systems. Heavy vehicles differ from light vehicles in the systems they comprise, the ways they are developed, and how they are used. The heavy-vehicle industry is applying the principles of functional safety through established systems engineering practices or the general industry standard IEC 61508. A revision of ISO 26262 that will expand the scope to include trucks and buses was in the committee draft stage at the time this study was conducted. The revision is expected to clarify the demarcation between a truck and attached vocational equipment and to account for the wider variance in heavy vehicles, but not to fundamentally change the process deriving the requirements for functional safety.

Baseline Analysis of Driver Performance at Intersections for the Left-Turn Assist (LTA) and Intersection Movement Assist (IMA) Applications

This study supports development of left turn assist (LTA) and intersection movement assist (IMA) applications that provide warnings to drivers crossing intersections. It supports improved intersection collision warning applications design by enhancing understanding of intersection behavior, identifying metrics and test procedures through analysis of real-world data, and providing information used to reduce false alerts and nuisance alerts. A literature review identifies previous research into metrics for driver behavior at intersections, and databases from two naturalistic driving studies were queried to identify scenarios that could be analyzed as examples of normal or baseline turning behavior. Crashes were also identified in a national crash database and analyzed as examples of driving where alert would have been useful.
 

Occupant Safety in Vehicles Equipped With Automated Driving Systems, Part 1: Initial Evaluation of Usability, Stability, and Injury Prediction Capabilities

This project sought to perform an initial evaluation of the usability, stability, and potential injury prediction capabilities for two human body models (GHBMC-M50-O and GHBMC-M50-OS) and the NHTSA THOR FE anthropomorphic test device model in occupant postures that will become more possible with automated driving systems. Postures examined degrees of seat recline, inboard seat rotation, occupants turned in their seats, and occupants leaning against the belts in sleep-like posture. Collision scenarios included moving deformable barrier impacts at PDOFs around the vehicle. All simulations were performed with the occupant seated in the right front passenger position. Restraints included a front passenger air bag, a side curtain air bag, a side torso air bag, and a 3-point seat belt. Approximately 175 full vehicle simulations were performed and analyzed.

Occupant Safety in Vehicles Equipped With Automated Driving Systems, Part 3: Biofidelity Evaluation of GHBMC M50-OS Against Laboratory Sled Tests

This project sought to perform an initial evaluation of the usability, stability, and potential injury prediction capabilities for two human body models (GHBMC-M50-O and GHBMC-M50-OS) and the NHTSA THOR FE anthropomorphic test device model in occupant postures that will become more possible with automated driving systems. Postures examined degrees of seat recline, inboard seat rotation, occupants turned in their seats, and occupants leaning against the belts in sleep-like posture. Collision scenarios included moving deformable barrier impacts at PDOFs around the vehicle. All simulations were performed with the occupant seated in the right front passenger position. Restraints included a front passenger air bag, a side curtain air bag, a side torso air bag, and a 3-point seat belt. Approximately 175 full vehicle simulations were performed and analyzed.

Parameter Study of the OMDB Test Procedure

Oblique impact configurations account for a significant amount of real-world accidents. Compared to frontal crashes, these have different occupant kinematics and vehicle intrusion patterns. Consequently, a new oblique impact test is being developed and investigated by NHTSA. For example, offset moving deformable barrier (OMDB) impact velocity and occupant seating position can only be controlled within certain limits. Finite element simulations, consisting of detailed computer models of a vehicle, the OMDB, the THOR crash test dummy, and relevant restraints and interiors were used. Advanced design of experiment methods were applied to determine the importance of parameters and their effect on the vehicle and occupant criteria.