Potential Reduction in Crashes, Injuries and Deaths from Large-Scale Deployment of Advanced Driver Assistance Systems

A synthesis of existing research on the potential safety benefits of selected Advanced Driver Assistance Systems (ADAS) providing new estimates of the numbers of crashes, injuries, and deaths that such systems could potentially help prevent.
   This research brief reviews recent literature and provides updated statistical estimates regarding the numbers of crashes, injuries, and deaths that could theoretically be addressed by equipping all cars, pickup trucks, vans, minivans, and sport utility vehicles with advanced driver assistance systems (ADAS) technologies. Technologies included in the scope of this brief are designed to prevent or reduce the severity of specific types of crashes, or to help the driver do so. Specific technologies examined are: forward collision warning (FCW), automatic emergency braking (AEB), lane departure warning (LDW), lane keeping assistance (LKA), and blind spot monitoring (BSM) systems. Driver assistance technologies designed primarily for driver convenience (e.g., adaptive cruise control systems; parking assistance systems) are outside the scope of this review. This brief estimates the numbers of crashes, injuries, and deaths that these technologies could theoretically help prevent or mitigate, it does not attempt to quantify the likely actual real-world reductions in crashes, injuries, and deaths attributable to these technologies.
Key Findings
The technologies examined are estimated to have the potential to prevent a combined total of approximately 40% of all passenger-vehicle crashes, 37% of injuries that occur in crashes involving passenger vehicles, and 29% of all deaths in crashes that involve passenger vehicles.
   FCW/AEB and LDW/LKA systems were each estimated to have the potential to help prevent approximately 14% of all motor vehicle crash fatalities. However, FCW/AEB systems were estimated to be relevant to more than four times as many crashes and injuries as LDW/LKA. This is because the types of crashes targeted by LDW/LKA systems, i.e., lane departure crashes and especially single-vehicle road departure crashes, tend to be more severe than most other crash types. FCW and AEB systems with pedestrian detection capability have the potential to prevent a substantial number of fatalities, especially involving pedestrians and cyclists; however, most of the overall crashes to which they are relevant are rear-end crashes, which are rarely fatal. The overall contribution of BSW systems to crash reductions was the smallest by all measures, but they are still estimated to have the potential to help prevent as many as 318,000 crashes annually.
   This study sought to quantify the number of crashes, injuries, and deaths that occurred in the United States in 2016 that theoretically might have been avoided or reduced in severity if all vehicles had been equipped with the technologies of interest. Data on police-reported crashes and injuries that occurred in crashes were from the NHTSA’s Crash Report Sampling System, which comprises a representative sample of all police-reported motor vehicle crashes nationwide. Data on deaths that occurred in crashes were from the NHTSA’s Fatality Analysis Reporting System, which is a census of all fatal motor vehicle crashes nationwide.
   Variables pertaining to the vehicle, driver, geometry of the crash, environmental conditions, and the sequence of events of the crash were used to identify crashes that the systems of interest are designed to prevent or mitigate. Crashes of potential interest in the current analysis were those that involved a passenger vehicle in the first harmful event of the crash.
   First, researchers examined the data to identify crashes of the general type that each respective technology is designed to address (e.g., LDW systems are designed to prevent crashes in which the vehicle leaves its travel lane unintentionally). Researchers then identified and subtracted specific subsets of those crashes that the technology likely would not have prevented due to its known limitations (e.g., sensors not functioning reliably in inclement weather) or idiosyncratic factors present in the crash (e.g., the driver was intoxicated and thus might not have responded appropriately to warnings). The remaining crashes represent crashes that theoretically might have been prevented if the vehicles involved in the crash had been equipped with the technologies examined.