Quantifying the Lost Safety Benefits of ADAS Technologies Due to Inadequate Supporting Road Infrastructure

Abstract

Advanced driver assistance systems (ADAS) provide warnings to drivers and, if applicable, intervene to mitigate a collision if one is imminent. Autonomous emergency brakes (AEB) and lane keep assistance (LKA) systems are mandated in several new vehicles, given their predicted injury and fatality reduction benefits. These predicted benefits are based on the assumption that roads are always entirely supportive of ADAS technologies. Little research, however, has been conducted regarding the preparedness of the road network to support these technologies in Australia, given its vastly expansive terrain and varying road quality. The objective of this study was to estimate what proportion of crashes that are sensitive to AEB and LKA, would not be mitigated due to unsupportive road infrastructure, and therefore, the lost benefits of the technologies due to inadequate road infrastructure. To do this, previously identified technology effectiveness estimates and a published methodology for identifying ADAS-supportive infrastructure availability was applied to an estimated AEB and LKA-sensitive crash subset (using crash data from Victoria, South Australia and Queensland, 2013–2018 inclusive). Findings demonstrate that while the road networks across the three states appeared largely supportive of AEB technology, the lack of delineation across arterial and sub-arterial (or equivalent) roads is likely to have serious implications on road safety, given 13–23% of all fatal and serious injury (FSI) crashes that occurred on these road classes were LKA-sensitive. Based on historical crash data, over 37 fatalities and 357 serious injuries may not be avoided annually across the three Australian states based on the lack of satisfactory road delineation on arterial and sub-arterial (or equivalent) roads alone. Further, almost 24% of fatalities in Victoria, 24% of fatalities in Queensland and 21% of fatalities in South Australia (that are AEB- or LKA-sensitive) are unlikely to be prevented, given existing road infrastructure. These figures are conservative estimates of the lost benefits of the technologies as they only consider fatal and serious injury crashes and do not include minor injury or property damage crashes, the benefits of pedestrian-sensitive AEB crashes in high-speed zones or AEB fitted to heavy vehicles. It is timely for road investments to be considered, prioritised and allocated, given the anticipated penetration of the new technologies into the fleet, to ensure that the road infrastructure is capable of supporting the upcoming fleet safety improvements.

Too close for comfort: A mixed methods study to understand self-reported tailgating using the theory of planned behaviour

“…Adaptive cruise control systems also assist drivers in keeping a safe headway. However, these will also be reliant upon appropriate infrastructure (Peiris et al, 2022).…”

Automated Lane Centering: An Off-the-Shelf Computer Vision Product vs. Infrastructure-Based Chip-Enabled Raised Pavement Markers

“…Lastly, the third test route was a portion of the WMU campus drive with a low curvature. It has been observed that there is a notable difference in the performance of vehicle automation products on arterial roads with steep road curvatures, uneven road surface finish, improper lane markings, and lane marking obstructions [30]. The CERPM technology can be tested to address these drawbacks.…”

ADAS: Should we be too Reliant for Driver Safety?

“…As an example, it is worth noting that forwardcollision warning systems encounter challenges in detecting objects while a vehicle is in motion at elevated velocities. However, four of these technologies are particularly noteworthy in terms of the difficulties they pose, namely processors, sensors, software algorithms, and mapping [3][4][5][6].…”