Despite the fact that semi-autonomous vehicles will become more and more prevalent in the coming decades, recent studies have highlighted that traffic accidents will persist as a core issue for road users, insurers, and policy makers alike. Researchers and industry players see potential in the technology embedded in semi-autonomous vehicles to combat this challenge by reliably predicting locations with a high likelihood of traffic accidents. This technology can be leveraged to detect accidents and 'near miss incidents', such as heavy braking and evasive manoeuvres, otherwise known as Critical Driving Events (CDEs). The locations of CDEs could identify areas of high accident exposure, offering automotive insurers a unique opportunity to reduce traffic accidents through the adoption of active loss prevention business models, such as providing saferouting services and in-vehicle warnings. To date, there is limited empirical evidence on whether the Crash Frequency and Crash Rate of locations can be accurately identified through CDEs. To address this research gap, an 18-week naturalistic driving field study of 72 vehicles was conducted in Switzerland, covering over 690,000 km. Data collected from the CAN Bus of these vehicles indicate that there is a proportional relationship between the CDEs of the fleet, and the Crash Frequency and Crash Rate of a location. Furthermore, a nationwide spatial regression analysis was applied to determine Crash Frequency across the majority of the Swiss road network. We identify the relationship between Crash Frequency, and the CDEs and Trip Frequency of the fleet, along with additional explanatory variables for urban and highway locations. These insights provide first evidence that insurance companies and other industry players with access to a nationwide semi-autonomous fleet can determine existing and emerging locations of high accident probability, enabling more proactive business models and safety focused services.
A B S T R A C TDespite continuous investment in road and vehicle safety, as well as improvements in technology standards, the total amount of road traffic accidents has been increasing over the last decades. Consequently, identifying ways of effectively reducing the frequency and severity of traffic accidents is of utmost importance. In light of the depicted challenge, latest studies provide promising evidence that in-vehicle decision support systems (DSSs) can have significant positive effects on driving behaviour and collision avoidance. Going beyond existing research, we developed a comprehensive in-vehicle DSS, which provides accident hotspot warnings to drivers based on location analytics applied to a national historical accident dataset, composed of over 266,000 accidents. As such, we depict the design and field evaluation of an in-vehicle DSS, bridging the gap between real world location analytics and in-vehicle warnings. The system was tested in a country-wide field test of 57 professional drivers, with over 170,000km driven during a four-week period, where vehicle data were gathered via a connected car prototype system. Ultimately, we demonstrate that in-vehicle warnings of accident hotspots have a significant improvement on driver behaviour over time. In addition, we provide first evidence that an individual's personality plays a key role in the effectiveness of in-vehicle DSSs. However, in contrast to existing lab experiments with very promising results, we were unable to find an immediate effect on driver behaviour. Hence, we see a strong need for further field experiments with high resolution car data to confirm that in-vehicle DSSs can deliver in diverse field situations.
Driver identification is a growing topic which offers a streamlined user experience in the connected car, but potentially also highlights privacy issues of our interconnected lives. Recent studies have reported the ability for individuals to be reliably identified out of a group based on their driving behavior. In particular, the state-of-the-art study claims that, in a controlled setting, data collected on how a driver operated the brake pedal could perfectly distinguish between 15 drivers. The paper at hand was not able to validate these strong scientific claims using naturalistic driving data. In line with the results of other studies using similar data, the replicated identification accuracy dropped to values between 40% and 70% by applying the outlined methods. Nevertheless, this paper further contributes by adapting the reported feature collection technique in order to achieve identification results between 80% and 99.5% in this challenging setting, thus advancing the state-of-the-art. These findings demonstrate the real-world capabilities of data-enabled driver identification, which both facilitates new use-cases and potentially raises privacy questions. As such, important key features from the identification models are presented to assist both researchers and practitioners in this rapidly developing topic.
About 17% of the worldwide CO2-emissions can be ascribed to road transportation. Using information systems (IS)enabled feedback has shown to be very efficient in promoting a less fuel-consuming driving style. Today, in-car IS that provide feedback on driving behavior are in the midst of a fundamental change. Increasing digitalization of in-car IS enables virtually any kind of feedback. Still, we see a gap in the empirical evidence on how to leverage this potential, raising questions on future HCI-based feedback design. To address this knowledge gap, we designed an eco-driving feedback IS and, building upon construal level theory, hypothesize that abstract feedback is more effective in reducing fuel consumption than concrete feedback. Deployed in a large field experiment with 56 participants covering over 297,000km, we provide first empirical evidence that supports this hypothesis. Despite its limitations, this research may have general implications for the design of real-time feedback.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
hi@scite.ai
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.