ABSTRACT
This article presents a synthetic distracted driving (SynDD1) dataset for machine learning models to detect and analyze drivers' various distracted behavior and different gaze zones. We collected the data in a stationary vehicle using three in-vehicle cameras positioned at locations: on the dashboard, near the rearview mirror, and on the top right-side window corner. The dataset contains two activity types: distracted activities [1], [2], [3], and gaze zones [4], [5], [6] for each participant and each activity type has two sets: without appearance blocks and with appearance blocks, such as wearing a hat or sunglasses. The order and duration of each activity for each participant are random. In addition, the dataset contains manual annotations for each activity, having its start and end time annotated. Researchers could use this dataset to evaluate the performance of machine learning algorithms for the classification of various distracting activities and gaze zones of drivers.
ABSTRACT
Bicyclists are vulnerable road users who are at a greater risk for injury and fatality during crashes. Additionally, the "near-miss" incidents they experience during regular trips can increase the perceived risk and deter them from riding again. This paper aims to use naturalistic bicycling data collected in Johnson County, Iowa to: 1) study the effect of factors such as road surface type, parked vehicles, pavement markings and car passing events on cyclists' physiological stress and 2) understand the effect of daytime running lights (DRL) as an on-bicycle safety system in providing comfort to cyclists and highlight of their presence on the road to other vehicles. A total of 37 participants were recruited to complete trips over two weekends, one weekend with DRL and the other without DRL. Recruitment was specifically targeted toward cyclists who expressed discomfort riding in traffic. Data were collected using a front forward facing camera, GPS, and a vehicle lateral passing distance sensor mounted on the bicycle and a Empatica E4 wrist band (providing physiological data such as electrodermal activity; EDA) worn by the cyclist. Data from those sources were cleaned, processed, merged, and aggregated into time windows depicting car passing and no car passing events. Mixed effects models were used to study the cyclists' skin conductance response (phasic EDA) and baseline skin conductance level (tonic EDA). Car passing, parked vehicles, and roads with dashed centerline markings were observed to increase the cyclists stress. The use of DRL had negligible impact on cyclist stress on roads.
