As the world moves towards higher levels of vehicle automation, the interplay between human drivers and automated systems becomes increasingly complex. This paper addresses the challenges of navigating the intermediate levels of vehicle automation, the transition stage from human driven vehicles to automated vehicles. The step-by-step introduction of automated features introduces new risks, such as mode confusion and over-reliance on automation, as well as mental underload or overload which lead to decreased driver performance and increased crash risk. In addition, during the transition, automation technology is still maturing and also has its limitations. In our view, we should aim to integrate the strengths of both human drivers and automation to enhance traffic safety and driver comfort. This paper aims to contribute conceptually to the scientific discourse on vehicle automation and to focus future research. It presents four key concepts that have proven to be meaningful to change perspective, including the Driver/Automation Fitness Plane, the definition of human-centered driving modes, and the mediator approach to seamless collaboration between driver and automation. These concepts are designed to facilitate a safer and more intuitive interaction between humans and automated systems, leveraging interdisciplinary perspectives on technology, behavior, cognition, and design. The paper concludes with a discussion on the potential of automation and calls for a human-centered approach to fully realize the benefits of vehicle automation.

Within the UDRIVE project, a rich cross-European naturalistic driving database was created which includes everyday driving data on car and truck drivers and powered two-wheeler riders. The database provides extensive, reliable insights into driving behavior in real traffic as a foundation for improving the safety and sustainability of European road traffic. This paper discusses the characteristics of the data in the UDRIVE database—elucidating key methodological choices and presenting a selection of results to date. A priority of the study design was obtaining in-depth information on driving behavior, permitting the exploration of diverse research questions. A tailor-made data acquisition system collected very comprehensive data. A total of 287 drivers/riders participated. The sample size restricts the addressable research topics to common behaviors in everyday driving and limits the generalizability of results. However, the data are extensive and promising analyses have already been performed. The results show differences between European countries for distracting activities, seatbelt use, and looking behavior towards cyclists at urban intersections. Moreover, it shows that European drivers engage less in mobile phone use than U.S. drivers. It is likely that European drivers differ in other ways, also—highlighting the dataset's value for developing and implementing targeted safety measures, for the E.U. and its individual countries. Based on the comparison of the different studies, the paper introduces the general conceptual framework for naturalistic driving studies, providing insight in the relation between the scope of a naturalistic driving study and the key methodological choices on sample selection and data acquisition system.

The transition period towards large-scale or full automated driving will pose specific challenges. One of the challenges concerns the interaction of automated vehicles with vulnerable road users. So far, most studies into this type of interactions took the perspective of the car. The current study, however, takes the perspective of the vulnerable road user. More specifically, it explores how cyclists perceive and expect automated vehicles to ‘behave’ and how cyclists would react. Expectations are important determinants of traffic behaviour. Incorrect expectations could lead to overly trustful or hesitant behaviour and subsequent unsafe interactions. Recently, Hagenzieker and colleagues (2017) conducted a photo experiment in which regular cyclists had to judge photos of traffic situations where they encountered manually-driven cars and automated cars (recognisable by either a sticker on the side of the car or a roof name plate on top of the car). Participants judged 30 photos twice, in random order. In that study a subset of nine participants were equipped with an eye tracker in order to study their eye movements while judging the photos. This study further examined these eye tracking data, comparing timeto- first-fixation, dwell time from the start of the first fixation, and total number of revisits in interactions with automated and with manually-driven cars. Results indicate no differences in time-to-first-fixation nor in the number of revisits between situations with automated cars and traditional cars. Dwell times revealed an effect of familiarity, showing that cyclists spent more time looking at cars during the first round of photos compared to the second round. In particular, they spent more time looking at automated cars which were identifiable by a ticker on the side. The results are discussed and suggestions for future research are given.