Improving Pedestrian Trust in Autonomous Vehicles: Implications for AV Interface Design From an Online Study Jack Neuhoff, Jinjuan She This research aims to understand communication between pedestrians and autonomous vehicles (AV) to inform AV interface design. Effective communication is communication that results in high levels of pedestrian trust in an AV and decision alignment with the AV’s expectations, to ensure that pedestrians can cross the road safely while feeling safe and comfortable. High levels of trust lead to increased acceptance of new technologies. Our project started with identifying existing communication methods between AVs and pedestrians. Two significant forms of communication were found between pedestrians and AVs, which consisted of implicit and explicit communication. Implicit communication consists of revealing intention through action, for example slowing down at a cross walk communicated intent to yield right of way. Explicit communication is more direct, involving a signal, sound, etc. We then categorized each explicit communication method into three styles: informative, advisory, and commanding. Additionally, the use of adaptive communication was analyzed to determine its influence on pedestrian trust. Adaptive communication is communication that is initiated based on some level of feedback. For example, if an AV explicitly communicated intent to yield to a pedestrian and they seem unsure or hesitant, a second message may be used to further reassure them it is safe to cross. To test the effectiveness, we conducted an online human subject study with 223 participants. The study consisted of a simple scenario: a pedestrian is about to cross a 2-way street on a crosswalk with no stop sign or flashing signals and the pedestrian relies on the approaching vehicle’s communication and the scenario to make a decision. &nbsp;Pedestrians’ level of trust with each communication style was measured. In some instances, when pedestrians seemed unsure of whether to cross or not, adaptive communication was utilized in order to determine if it had a significant effect on trust. The data was analyzed with Linear Mixed Model (LMM). It is found that the advisory (Mean±SE: 6.37±0.15) and commanding style (Mean±SE: 6.56±0.15) communication methods are more effective in increasing pedestrian trust, than informative styles (Mean±SE: 6.11±0.16). Furthermore, the addition of adaptive communication was shown to significantly improve level of trust in AVs (Mean±SE of adaptive communication : 6.52±0.12 vs. Mean±SE of single communication : 5.51± 0.13, p-value&lt;2e-16). The results suggest that an AV’s communication style plays a crucial role in affecting pedestrians’ trust in the AV. Additionally, it is important to incorporate adaptive communication strategy into AVs in order to improve pedestrians’ trust and thus make cross decisions as the AV expects. Both of these outcomes help AV researchers and designers better understand and consider pedestrians’ needs during interaction and interface design. In the long run, this research could further enable us to redesign more intelligent AV user interfaces that are capable of adjusting their communication strategies based on the response from human and keeping good communication dynamics.


Improving Pedestrian Trust in Autonomous Vehicles: Implications for AV Interface Design From an Online Study

Division / Sponsors: 
Aerospace, Bioengineering, Applied Mechanics, Design Engineering Division, Advanced Energy Systems, Manufacturing Engineering Division, Noise Control and Acoustics, Nondestructive Evaluation, Diagnosis, and Prognosis (NDPD), Nuclear Engineering Division, Process Industries Division, Fluids Engineering, Heat Transfer, Materials, Microelectromechanical Systems, Safety Engineering and Risk Analysis, Computers & Information in Engineering


Paper Proceedings: https://asmedigitalcollection.asme.org/IMECE/IMECE2020/volume/84478

IMECE2020 - Virtual Conference

Supporting the need among researchers worldwide for additional online and multimedia content, ASME now offers the ASME Conference Video Collection. This platform will collect and provide long-term access to conference presentation videos, encompassing all technology areas within ASME's technical conference and events program. About 30,000-40,000 engineers across all mechanical engineering and related disciplines participate in ASME conferences each year. This includes authors of approximately 6,000 to 8,000 technical papers published by ASME, the majority of whom also submit video presentations for our virtual conferences. The video collection will be available as an add-on to existing product packages or as a separate subscription.