东南大学学报（英文版）2021，Vol.37Issue(3)：325-331,7.DOI:10.3969/j.issn.1003-7985.2021.03.013
城市隧道驾驶员注视行为特征
Drivers' fixation transfer characteristics in urban tunnels
摘要
Abstract
To improve the safety performance of urban tunnels,the fixation transfer characteristics of drivers with different driving experience levels in urban tunnels were investigated.First,a real vehicle test was performed in an urban tunnel,and the eye movement data of 10 drivers with different driving experience levels were collected using a Dikablis eye-tracking system.Second,the driver fixation range was divided into eight areas of visual interest by using the K-means clustering method,and the fixations in different sections of the tunnel were comparatively analyzed.Finally,on the basis of the divided areas of visual interest,fixation transfer rules and the stationary distribution characteristics of drivers with different driving experience levels on different sections of the tunnel were discussed using Markov theory.Results indicate that drivers' probability of repeated fixation is greater and that the efficiency of visual search is lower at internal sections of tunnels than in external sections.Drivers obtain information mainly from the straight upper from and straight lower front areas,and the probabilities of fixation points in these two areas at the threshold and exit sections are significantly higher than those in other sections.Relative to experienced drivers,novice drivers allocate little attention to the straight upper front area and rear-view mirrors.Hence,they have weak fixation when looking forward,and they lack experience in obtaining information on rear-approaching vehicles and controlling speed.关键词
交通安全/城市隧道/眼动/固定传递/马尔可夫理论Key words
traffic safety/urban tunnel/eye movement/fixation transfer/Markov theory分类
交通工程引用本文复制引用
郭唐仪,潘姝,邵飞,徐倩..城市隧道驾驶员注视行为特征[J].东南大学学报（英文版）,2021,37(3):325-331,7.基金项目
The National Key Research and Development Pro-gram of China (No.2019YFE0123800),the Fundamental Research Funds for the Central Universities (No.30919011290,30920010010). （No.2019YFE0123800）
