The deliverable “D7.2 L3/L4 long-term study about user experiences” describes work within L3Pilot that investigated the change of user experience with long-term usage of an ADF. In the literature, this topic is summarised under the broader concept of behavioural adaptation. During the project it turned out that there are unanswered questions regarding behavioural adaptation in addition to the question of system perception that is built on repeated exposure. Due to that, the scope of this deliverable has been broadened to studies conducted in L3Pilot that deal with behavioural adaptation independent of the considered time span and closely related user topics.
The deliverable starts with an overview of reported research and a short look at the literature. Then, eight chapters summarise eight supplementary studies which addressed user related topics by using methods that go beyond the on-road pilot testing in L3Pilot. Three studies investigated the change of acceptance and usage with repeated usage of a motorway chauffeur. One study took place in a driving simulator, one used a Wizard of Oz vehicle and one addressed the topic as part of the on-road pilot testing. Furthermore, there is work focusing on short-tem behavioural adaptation, i.e. an immediate impact of driving with an ADF on manual driving behaviour directly after or during the transition of control. Lastly, drivers’ preferences with regard to potential side tasks while driving with an automated driving system were explored in an online-survey.
Overall, it was shown that when using an automated driving functions on multiple drives, drivers get used to the system, they gain trust and evaluate the function positively. The studies on shortterm behavioural adaptation indicate that further research is needed to better understand whether and how driving with an automated driving function impacts manual driving behaviour immediately after transferring control back to the driver.
For future work it is recommended to further investigate the topic of behavioural adaptation in the long- but also the short-term perspective. This will help to design automated driving systems in way that drivers feel like using the function and that usage is safe right from the beginning and stays safe after getting used to the system.
