CONTACT US: Contact info
|
Citation |
Woide M, Stiegemeier D, Baumann M. Proc. Int. Driv. Symp. Hum. Factors Driv. Assess. Train. Veh. Des. 2019; 2019: 314-320. |
|
Copyright |
(Copyright © 2019, University of Iowa Public Policy Center) |
|
DOI |
unavailable |
PMID |
unavailable |
|
Abstract |
Future autonomous vehicles will make their own maneuver decisions whereby situations will occur in which the maneuver performed by the autonomous vehicle contradicts the course of action preferred by the driver. In response, the uninformed driver takes over manual control of the vehicle and performs a potentially inappropriate and safety-critical maneuver due to a lack of information. To prevent such a behavior in future, a methodical paradigm is needed, which is able to create possible driver - autonomous vehicle - conflicts and examine preventive and cooperative solutions in a driving simulator. This study (n = 29) is a successful methodical approach to create possible, authentic and reproducible driver - autonomous vehicle - conflicts. Conflicts were caused by a combination of gradation of visibility by fog (full visibility, 150m, 100m, 50m) and a maneuver performed by the automation (overtaking, following) on a rural road. 83% of the drivers canceled an overtaking maneuver by the automation and took over manual control in the 50m condition compared to 2% in the full visibility condition (z=1.914, p<.00). If the automation performed a following maneuver in the full visibility condition, 95% of the drivers overtook manually, compared to only 6% at 50m visibility (z=2.069, p<.00). Language: en |