Search Results

Pedal errors have been widely reported as a leading cause of unintended acceleration (UA) incidents for several decades. Many governmental and scientific studies have attempted to characterize the rate of pedal errors leading to UA incidents using data from the North Carolina Crash Database. These data, however, are limited for various reasons, including the absence of an in-depth investigation of causal factors contributing to the accident. To further examine the rate of UA incidents related to pedal error, we utilized the National Motor Vehicle Crash Causation Survey (NMVCCS), a nationally representative sample of 5,471 crashes that occurred between 2005 and 2007. Using a targeted keyword search, we identified 48 potential pedal errors (30 driver-admitted), providing a national estimate of 17,919 pedal errors. We then investigated accident characteristics across these specific cases, including demographics of the drivers, vehicle characteristics, and pre-crash critical events.

Unintended acceleration events due to pedal misapplication have been shown to occur more frequently in older vs. younger drivers. While such occurrences are well documented, the nature of these movement errors is not well-characterized in common pedal error scenarios: namely, on-road, non-emergency stopping or slowing maneuvers. It is commonly assumed that drivers move in a ballistic or “direct hit” trajectory from the accelerator to the brake pedal. However, recent simulator studies show that drivers do not always move directly between pedals, with older drivers displaying more variable foot trajectories than younger drivers. Our study investigated pedal movement trajectories in older drivers ages 67.9 ± 5.2 years (7 males, 8 females) during on-road driving in response to variable traffic light conditions. Three different sedans and a pick-up truck were utilized.

The usage of rearview camera displays and their effectiveness on drivers' capability to avoid unexpected obstacles during four common backing tasks (i.e., parallel parking, backing between two vehicles, backing down a driveway, backing out of a garage) was evaluated on a closed-course with stationary confederate vehicles, signage, and lane markings. The obstacle consisted of either a stationary or a moving target that appeared to the rear of the test vehicle. Eye movements and vehicle dynamics measurements (i.e., longitudinal acceleration, brake displacement) were recorded, in addition to obstacle hit/avoidance rates. Performance was assessed for four rearview camera (RVC) conditions: small center-stack display (SD), large center-stack display (i.e., navigation screen) (LD), in-mirror display (IMD), and no display (ND).

Sound localization of a backup alarm is important in situations when vehicles are reversing. Previous work has demonstrated the effects of ambient noise level and the spectral content of the backup alarm on localization. In the current study, we investigate the effects of backup alarm mounting location on localization performance. To address this question, we asked blindfolded listeners to localize backup alarms installed in positions that provided either direct (e.g., installed on the outer rear aspect of the vehicle) or indirect (e.g., installed within the inner frame rails of the vehicle) sound propagation paths to the listener. Additionally, we explored the effects of ambient noise level and the direction of origin of the alarm (behind, in front of, or to the left or right of the listener), and the interactions among all three factors (alarm location, ambient noise, and alarm direction relative to the listener).