Date of Graduation
Statler College of Engineering and Mineral Resources
Civil and Environmental Engineering
Public perceptions have been playing an important role in the development of autonomous vehicle (AV) technology. Besides AV and non-AV users, the perceptions of vulnerable roadway users are critical, as AVs will become a part of multimodal transportation system. Pedestrians and bicyclists are among the vulnerable groups of roadway users, as they are relatively unprotected compared to the occupants of AVs or non-AVs. Although AV’s capability to monitor other vehicles has been documented in many studies, there are concerns about AV’s capability in monitoring pedestrians and bicyclists. The overarching goal of this dissertation is to investigate the perceptions of pedestrians and bicyclists on AVs to understand and incorporate their perceptions in AV technology development. The specific research objectives are to- (i) categorize the positive and negative perceptions and regulation expectations of pedestrians and bicyclists, (ii) identify factors influencing AV road sharing related safety perceptions among pedestrians and bicyclists, (iii) understand pedestrians’ and bicyclists’ expectations on AV regulations and identify relevant factors influencing their attitudes towards AV regulations, and (iv) investigate the effectiveness of widely used close-ended rating-based quantitative survey question to assess AV perceptions among pedestrians and bicyclists. Two surveys conducted by Bike Pittsburgh (BikePGH) were used to accomplish the research objectives. In addition to quantitative responses, BikePGH surveys collected open-ended responses to understand the reasons for pedestrians’ and bicyclists’ quantitative responses.
A combined inductive and deductive qualitative data analysis approach was applied to classify pedestrians’ and bicyclists' positive and negative perceptions and regulation expectations. Pedestrians and bicyclists expressed comparatively fewer negative opinions towards AVs than positive opinions. Negative opinions included a lack of safety and comfort around AVs and trust in the AV technology. Respondents also concerned about AV technology issues (e.g., slow and defensive driving, disruptive maneuvers). Pedestrians’ and bicyclists’ opinions were significantly influenced by their views on AV safety, familiarity with the AV technology, exposure to AV-related news, and household automobile ownership. Regulating AV movement on public roadways, developing safety assessment guidelines, and controlling oversights of AV technology developers' improper practices were the survey participants' noteworthy suggestions.
Non-parametric statistical tests were conducted to compare the safety perceptions of pedestrians and bicyclists based on their characteristics, experiences, and attitudes. An ordered probit model was estimated to quantify the influence of different factors on safety perceptions of pedestrians and bicyclists regarding road sharing with AVs. In addition, safety perceptions and the effect of various factors on AV safety perceptions over time were evaluated. Exposure to AV operations was associated with improved safety perceptions. Respondents with a stricter attitude towards AV regulations perceived road sharing with AVs as less safe. Despite the differences in safety perceptions among pedestrians and bicyclists in 2017 and 2019, different factors had a similar effect on AV road sharing-related safety perceptions in both surveys.
This dissertation also evaluated pedestrians’ and bicyclists’ perceptions on six AV regulations –(i) capping AV speed limit, (ii) operating AVs in manual mode in sensitive areas, (iii) requiring two employees while AVs operating in autonomous mode, (iv) sharing non-personal data with proper authorities, (v) disclosing information of AVs limitations, capabilities, and real-world performances with proper authorities, and (vi) reporting all safety-related accidents with proper authorities. Parametric and non-parametric tests and binary logit models were applied to evaluate the perceptions of pedestrians and bicyclists on AV regulations. Strong support for AV-related data-sharing regulations was found among the pedestrians and bicyclists compared to the other AV regulations assessed in this research. Older respondents showed higher approval of AV testing on public roadways and less support for regulating AVs. AV technology familiarity and safe road sharing perceptions resulted in lower support for AV regulations.
The effectiveness of close-ended rating-based quantitative survey question to assess AV safety perception was evaluated by extracting actual AV safety perceptions of pedestrians and bicyclists from their responses to open-ended questions and comparing with their assigned AV safety rating. Wilcoxon signed rank tests identified significant difference in AV safety perception assessed by two types of survey questions, which highlights the inefficiencies of close-ended rating-based quantitative survey question to assess AV perceptions of pedestrians and bicyclists accurately.
The findings of this dissertation provide an in-depth understanding of pedestrians’ and bicyclists’ perceptions on AVs, which can help AV companies develop critical improvements addressing the expectations of vulnerable roadway users such as pedestrians and bicyclists. Policymakers can establish policy guidelines to ensure safe road sharing with AVs for pedestrians and bicyclists. In addition, research findings can be used to create informed AV regulations and develop policies that could improve pedestrians’ and bicyclists’ attitudes/perceptions on regulating AVs and promoting AV technology deployment.
Rahman, Md Tawhidur, "Understanding Perceptions of Vulnerable Roadway Users on Autonomous Vehicles" (2022). Graduate Theses, Dissertations, and Problem Reports. 11294.