The Needed Features of Connected and Automated Vehicles to Prevent Passenger Car Crashes Caused by Driving Errors

Abstract

Connected and automated vehicles (CAVs) can enhance traffic safety considerably. However, as CAVs are currently under development, the safety impact cannot be assessed directly. In this study, driver-managed passenger car crashes with fatalities in Finland were investigated qualitatively to evaluate the needed features of the CAVs to avoid these crashes. The focus was on single-car crashes and collisions between passenger cars, in which the immediate risk factor was a driving error (n = 48). Most of the analysed crashes (33 of 48) were due to loss of control with typically adverse weather or road conditions. To avoid these crashes, a CAV should be able to adjust its speed according to the conditions. In 13 of 48 crashes, the car was under control prior to the crash. A reliable capability to recognize other road users is an important CAV feature, because observational errors were common in these cases. In addition, communication between the vehicles could assist in avoiding intersection crashes and crashes caused by a sudden change in weather conditions. This study increases knowledge on crashes related to driving errors and the needed features of CAVs to avoid these crashes. In particular, CAVs’ feature to adjust the speed is important, because cases of loss of control in adverse weather or road conditions were typical events.

Assessing the Impact on Road Safety of Automated Vehicles: An Infrastructure Inspection-Based Approach

“…The methodology has been applied to the CoEXist project use cases, one of which is presented as an example (Use Case 3). The results show a general risk decrease according to the higher penetration of AVs, which is in line with many other results from other studies, both considering the penetration level and the technological advances [2][3][4]. The analysis highlights that some safety issues may arise with the introduction of specific AV driving logics.…”

“…Although worded differently, this sentence has been used many times in different contexts, including being used by then US Secretary of Transportation Anthony Renard Foxx at the annual Automated Vehicle Symposium in 2016 (held in San Francisco on 18-21 July 2016). Even if the introduction of AVs will undoubtedly improve many aspects of road safety, as also proven by some recent research both considering low levels [2] and high levels of automation [3,4], the aforementioned sentence is questionable for several reasons: it does not consider technology mistakes which automated vehicles will make, it does not account for the remarkable human ability to "invent" solutions when under pressure, which considerably limits the number of crashes concerning near misses, and it does not account for any role for infrastructure in causing crashes. Today, the latter often takes place because there is a human (the driver), who always ultimately takes the blame, but this will no longer be possible with automated vehicles.…”

Real-Time Suitable Predictive Control Using SPaT Information from Automated Traffic Lights

“…Typically, CAVs have better convenience, safety, and the capabilities of vehicle-to-infrastructure and vehicle-to-vehicle communications, which provide opportunities to suggest driving speeds. The authors in [2] provide an analysis of accidents that could have been avoided with the help of CAV in the real world. Discussions on the benefits of communication that can be utilized to maneuver more safely are put forward.…”

Performance of Autonomous Vehicles in Mixed Traffic under different Demand Conditions

“…According to World Health Organization (WHO), each year, around 1.3 million people lose their lives due to road accidents and 93% of fatalities are reported in low to middle-income countries [11]. Among the factors contributing to road crashes, human error is considered as leading, and studies have shown that human error was a contributory factor in 93% of the accidents [12], and most of the accidents were caused by loss of control by human drivers [13]. The discrete nature of AVs will be able to eliminate such human errors.…”