Development of a Quantitative Survey Method for Traffic Behavioral States Considering Behavioral Context
Japan Society for the Promotion of Science:Grants-in-Aid for Scientific Research Grant-in-Aid for Scientific Research (C)
Date (from‐to) : 2018/04 -2021/03
Author : Hironobu Hasegawa
This research project was conducted for the following two purposes: 1) to automatically and accurately detect the activity state of pedestrians using time-series sensor data including physiological indices collected by an eyeglass-type device (eye measurement, 3-axis acceleration sensor, and 3-axis angular rate sensor) and a smartwatch (GPS), and 2) to develop a technology to identify the causal object from the eye measurement results. Of these, we achieved satisfactory results for the former. In addition, we were able to achieve mostly the latter. In addition, we proposed a methodology to test the effect of factors on the dissimilarity between scanpaths representing eye movements using PERMANOVA, and found that there was a significant effect on the relationship between gazing behavior and familiarity/unfamiliarity with the road environment and holding a driving license.