The full text of this item is not available at this time because the student has placed this item under an embargo for a period of time. The Libraries are not authorized to provide a copy of this work during the embargo period, even for Texas A&M users with NetID.
Biosensing of Location-based Environmental Distress of Pedestrians for Walkable Built Environments
Abstract
Urban built environments often include many negative stimuli (e.g., broken houses and abandoned vehicles) that are linked with stress symptomatology among urban populations. Biosignals (electrodermal activity, gait patterns, and heart rate) can help assess pedestrian distress levels induced by negative stimuli by mitigating the measurement limitations of traditional self-reporting methods and field observations. Despite their potential, utilizing assorted biosignals patterns is still obscure and needs further analysis. In this context, the goals of this research are to (1) investigate empirical associations between single/multiple negative stimuli and biosignals in naturalistic ambulatory settings; (2) propose and validate novel data-driven ways (saliency detection technique and multimodal classification model) to mitigate the impact of uncontrollable confounding factors from outdoor environments; and (3) examine the usefulness of biosignals acquired from real-life settings to capture the distress of pedestrians. Data collection was conducted in two different experimental settings. First, controlled route data collection was conducted from three different types of land use areas, including university (n=31), commercial (n=9), and residential (n=30). Participants were requested to walk on predefined routes with wearable devices (an inertial measurement unit, a wristband-type wearable device, an action camera, and a smartphone). Second, daily trip data collection was conducted by 67 subjects for six months. Each subject was asked to wear a wristband-type wearable device and carry a smartphone during a daily trip for a 2-week period. Additionally, participants provided self-reported scores on negative stimuli that existed during their walking trips through a survey. The results show that (1) statistical outcomes, obtained from biosignals, revealed short-term physiological responses to negative stimuli; (2) the suggested saliency detection technique was effective in capturing prominent local patterns of biosignals; (3) combined features from biosignals and image-based data presented enhanced prediction accuracy; and (4) biosignals are effectual in capturing pedestrian distress in daily life settings. The outcome of this research is expected to provide unique evidence for employing crowdsourced biosignals from community residents and opportunities for advancing facility/infrastructure management and urban built environment assessment.
Citation
Kim, Jinwoo (2022). Biosensing of Location-based Environmental Distress of Pedestrians for Walkable Built Environments. Doctoral dissertation, Texas A&M University. Available electronically from https : / /hdl .handle .net /1969 .1 /197913.