Methodology

The methodology utilized relied on target baseline crashes obtained from national crash databases, system performance data from characterization track tests, and basic kinematic computer simulation of vehicle-pedestrian conflicts. General Estimates System (GES) and Fatality Analysis Reporting Systems (FARS) crash databases from 2011 and 2012 were queried to identify and characterize vehicle-pedestrian crashes and to establish baseline crash information.

Three production PCAM systems were tested for warning and AEB capabilities, and were characterized by various testing conditions (i.e., vehicle parameters, pedestrian parameters, and environmental conditions). These conditions were correlated to the baseline crash information to obtain system effectiveness measures for the production PCAM systems. The PCAM systems were further defined by three different system logic implementations:

1. AEB only systems,
2. Warning + first braking response between AEB or driver, and
3. Warning + best braking response between AEB and driver.

A Monte Carlo simulation model was used to estimate the crash probability of vehicle-pedestrian conflicts given specific initial conditions. The simulation model exercised general kinematic equations in conjunction with driver and system performance data to determine the probability of a crash and the resulting impact speeds given a crash. Kinematic equations were derived from previous research for two priority pre-crash scenarios
1) Vehicle going straight and pedestrian crossing the road
2) Vehicle going straight and pedestrian walking alongside the road with/or against traffic

The simulation reconstructed historical PCAM-addressable GES and FARS crashes and superimposes PCAM system and driver performance data to determine the outcome with PCAM intervention.

Findings

The analysis in this report estimates that the crash avoidance effectiveness of PCAM systems can reduce up to 5,000 annual vehicle-pedestrian crashes and 810 fatal vehicle-pedestrian crashes. These crashes account for 8 percent of crash population where light-vehicles strike a pedestrian in the first event and 24 percent of same crash types where fatalities were involved.