| dc.description.abstract | World Health Organization data have indicated that in 75% of countries worldwide, bicyclists accounted for 2–19% of crash deaths. This crash risk may hamper efforts to increase bicycle usage and take advantage of its low carbon emissions. Therefore, understanding the risks associated with bicyclists′ use of roads will likely reduce crashes and may facilitate the use of bicycles.
This research fitted the sensor and device inclouding one global positioning system, 3-axis accelerometer, two ultrasonic distance sensors, eight proximity switches, one variable resistor, and five car camera DVR black boxes into an urban-style bicycle. Then, Using this bicycle to collect 40 bicyclists′ riding behavior data. In order to observe the valuable data from the great amount of row data, we established the data processing method to reduce the data and filter out the worthless data.
Throught the data reduction method to find the valuable data, we investigated vehicle-related factors, road-related factors, and bicyclist-related factors influenced motorists′ decisions about initial passing distances and bicyclists′ behaviors after the motorists started to pass. We found that many facotrs including road factor, bicylists′ gender difference and passing vehicles affected the motorists′ initial passing distance. The present study demonstrated that the quasi-naturalistic riding method is capable of collecting rich data concerning bicyclists′ behaviors, which could potentially be utilized in various types of studies.
| en_US |