Projects

Abstract: This study examines the societal impacts of E-scooters on disadvantaged populations in Austin, Texas. Through a population distribution analysis, it compares E-scooter use opportunities and space intrusion burdens among four vulnerable groups. Minority populations experienced fewer E-scooter use opportunities, with a disproportionate wait time for disturbance resolution. Ten percent… read more 

Abstract: In this study, we utilized a random forest model to predict the “L” train’s daily ridership in the Chicago downtown area during the pandemic based on environmental, transportation, and COVID-19-related factors. The results indicated that the model accurately predicts ridership one month in advance. However, its accuracy degraded over… read more 

Abstract: This study examines how the COVID-19 pandemic affected public transit ridership in Austin, TX, utilizing data from the Capital Metropolitan Transportation Authority and the American Community Survey. Through multivariate clustering and geographically weighted regression, it identifies demographic and spatial factors influencing ridership declines. Results indicate that areas with older… read more 

Abstract: This study aims to identify factors influencing e-scooter injury accidents in Austin due to concerns about rising ridership and insufficient accident data. Using 2018 dockless e-scooter injury data, we employed zero-inflated Poisson (ZIP) and zero-inflated negative binomial (ZINB) models. Results indicate the ZIP model better fits the data. Significant… read more 

Abstract: This study investigates the impact of ride-hailing on transportation access, particularly in low-density areas. Using data from Austin, Texas, we analyze ride-hailing usage, transit availability, and vehicle ownership across neighborhoods with varying demographics. Our findings reveal that ride-hailing has become an alternative mode of transport for residents in low-income,… read more 

Abstract: This study examines electric scooter (e-scooter) markets in U.S. transit deserts and oases, focusing on Austin, Chicago, Portland, and Minneapolis. Through t-tests, we compared e-scooter ride frequencies in these areas. Results show no significant difference in ride numbers between transit deserts and oases in Austin, Chicago, and Portland. Transit… read more 

Abstract: In this study, we explore graceful robot navigation in shared spaces, focusing on human-robot dyads resembling a dog and its handler. We examine five conditions: “Fully-Autonomous,” “Remote-Controlled,” “Companion,” “Leading,” and “Guided.” Participants observe these interactions and provide feedback via questionnaires. While initial questionnaire results show few significant differences, comparing… read more 

Abstract: Conventional U.S. rental housing data sources like the American Community Survey and American Housing Survey primarily capture transacted market data, reflecting existing renters’ payments. However, they do not directly reflect spot market conditions—the asking rents for current homeseekers. This study contrasts governmental data with millions of contemporary rental listings,… read more 

Abstract: The Longitudinal Social Impacts of HRI over Long-Term Deployments Workshop convenes researchers delving into various facets of understanding extended human-robot interaction deployments. Encompassing longitudinal studies, autonomy in prolonged contexts, and real-world implementations, the workshop aims to advance comprehension of how deployed robot systems influence individuals and societal dynamics. With… read more 

Abstract: Expanding the deployment of robots in social environments necessitates safe navigation in contact-prone settings. While collision-free navigation is well-studied, incorporating safe contacts remains underexplored. Traditional approaches mandate robots to freeze upon detecting imminent collisions, risking harm and impeding movement in dense crowds. To address this, we propose a learning-based… read more