Good User Interfaces for Automated Driving Continues to be Necessary

Best paper award for scientists Dr Lewis Chuang and Ms. Christiane Glatz from the Max Planck Institute of Biological Cybernetics in Tübingen

Best paper award went to the scientists Dr Lewis Chuang and Ms. Christiane Glatz from the Max Planck Institute of Biological Cybernetics in Tübingen in collaboration with Dr Stas Krupenia from Scania CV at AutomotiveUI 2017 in Oldenburg. Their studies were carried out in the Transregional Collaborative Research Centre (SFB-TRR 161), that also connects research groups at the Universities of Stuttgart and Konstanz developing quantitative methods for visual computing. 

The work of a truck driver can be physically and mentally exhausting. According to an official report by the Analysis Division of the Federal Motor Carrier Safety Administration, in 2015 alone 415,000 crashes were attributed to trucks on public traffic ways in the USA [1]. Vehicle automation can help to alleviate task demands and improve driving safety, thus providing drivers some relief from the tedium of driving and freeing them up to focus on more important jobs, such as logistics and route planning. Nonetheless, good user interfaces continue to be necessary, in order to ensure clear and transparent communications between the driver and the automated vehicle at all times. One example is the use of auditory notifications to warn drivers of potential collisions or to assist drivers in route planning.

Designing and implementing automotive user interfaces is a laborious process. Behavioural results that are demonstrated in the laboratory during user testing are rarely identical when implemented in more complex settings that resemble the real-world: One might be faster in responding to auditory warnings in the laboratory and much slower in the real world. Such differences are often difficult to reconcile.

Dr. Lewis Chuang and Ms. Christiane Glatz collaborated with Dr Stas Krupenia (Scania CV) to compare the brain activities of student volunteers in a sterile laboratory environment to those of professional truck drivers in a highly immersive truck simulator. In particular, they assessed how these individuals and their brains responded to auditory notifications that were designed to support communications between truck drivers and their automated vehicle.

In their paper, which was presented at the ACM International Conference for Automotive User Interfaces and Applications 2017, they reported that the brain activity (i.e. electroencephalography "EEG") of student volunteers and professional truck drivers were similar. Although professional truck drivers were found to be significantly slower and less accurate than student volunteers in responding to these auditory notifications, brain activity measurements revealed that the brains of professional truck drivers were no different from the students in processing these notifications for their presented information. "Behavioural responses to the same presented information can differ across different settings and different groups of individuals for many reasons," observes Dr Lewis Chuang. "Sometimes, it is because these notifications are no longer detectable or detectable in more complex environments. Other times, it is due to more mundane factors, for example, the possibility that truck drivers are generally older and perhaps more cautious than the average undergraduate student," and continues,"by knowing that the human brains of different individuals are responding to the same notifications in the same way, across different test settings, we are more confident that these auditory notifications were effective in communicating the information that they were intended for."

Printable images can be obtained at the Public Relations Office. Please send a proof upon publication.

Author: Beate Fuelle